Published as a conference paper at ICLR 2024

T IME M IXER : D ECOMPOSABLE M ULTISCALE M IXING

FOR T IME S ERIES F ORECASTING

Shiyu Wang1∗, Haixu Wu2∗, Xiaoming Shi1 , Tengge Hu2 , Huakun Luo2 , Lintao Ma1B ,
James Y. Zhang1 , Jun Zhou1B
1
Ant Group, Hangzhou, China 2 Tsinghua University, Beijing, China
{weiming.wsy,lintao.mlt,peter.sxm,james.z,jun.zhoujun}@antgroup.com,
{wuhx23,htg21,luohk19}@mails.tsinghua.edu.cn
arXiv:2405.14616v1 [cs.LG] 23 May 2024

A BSTRACT

Time series forecasting is widely used in extensive applications, such as traffic

planning and weather forecasting. However, real-world time series usually present
intricate temporal variations, making forecasting extremely challenging. Going
beyond the mainstream paradigms of plain decomposition and multiperiodicity
analysis, we analyze temporal variations in a novel view of multiscale-mixing,
which is based on an intuitive but important observation that time series present
distinct patterns in different sampling scales. The microscopic and the macroscopic
information are reflected in fine and coarse scales respectively, and thereby complex
variations can be inherently disentangled. Based on this observation, we propose
TimeMixer as a fully MLP-based architecture with Past-Decomposable-Mixing
(PDM) and Future-Multipredictor-Mixing (FMM) blocks to take full advantage of
disentangled multiscale series in both past extraction and future prediction phases.
Concretely, PDM applies the decomposition to multiscale series and further mixes
the decomposed seasonal and trend components in fine-to-coarse and coarse-to-fine
directions separately, which successively aggregates the microscopic seasonal and
macroscopic trend information. FMM further ensembles multiple predictors to
utilize complementary forecasting capabilities in multiscale observations. Conse-
quently, TimeMixer is able to achieve consistent state-of-the-art performances in
both long-term and short-term forecasting tasks with favorable run-time efficiency.

1 I NTRODUCTION
Time series forecasting has been studied with immense interest in extensive applications, such as
economics (Granger & Newbold, 2014), energy (Martín et al., 2010; Qian et al., 2019), traffic planning
(Chen et al., 2001; Yin et al., 2021) and weather prediction (Wu et al., 2023b), which is to predict
future temporal variations based on past observations of time series (Wu et al., 2023a). However, due
to the complex and non-stationary nature of the real world or systems, the observed series usually
present intricate temporal patterns, where the multitudinous variations, such as increasing, decreasing,
and fluctuating, are deeply mixed, bringing severe challenges to the forecasting task.
Recently, deep models have achieved promising progress in time series forecasting. The representative
models capture temporal variations with well-designed architectures, which span a wide range of
foundation backbones, including CNN (Wang et al., 2023; Wu et al., 2023a; Hewage et al., 2020),
RNN (Lai et al., 2018; Qin et al., 2017; Salinas et al., 2020), Transformer (Vaswani et al., 2017;
Zhou et al., 2021; Wu et al., 2021; Zhou et al., 2022b; Nie et al., 2023) and MLP (Zeng et al., 2023;
Zhang et al., 2022; Oreshkin et al., 2019; Challu et al., 2023). In the development of elaborative
model architectures, to tackle intricate temporal patterns, some special designs are also involved in
these deep models. The widely-acknowledged paradigms primarily include series decomposition and
multiperiodicity analysis. As a classical time series analysis technology, decomposition is introduced
to deep models as a basic module by (Wu et al., 2021), which decomposes the complex temporal
patterns into more predictable components, such as seasonal and trend, and thereby benefiting
the forecasting process (Zeng et al., 2023; Zhou et al., 2022b; Wang et al., 2023). Furthermore,
∗
Equal Contribution. Work was done while Haixu Wu, Tengge Hu, Huakun Luo were interns at Ant Group.

1
Published as a conference paper at ICLR 2024

multiperiodicity analysis is also involved in time series forecasting (Wu et al., 2023a; Zhou et al.,
2022a) to disentangle mixed temporal variations into multiple components with different period
lengths. Empowered with these designs, deep models are able to highlight inherent properties of time
series from tanglesome variations and further boost the forecasting performance.
Going beyond the above mentioned designs, we further observe that time series present distinct
temporal variations in different sampling scales, e.g., the hourly recorded traffic flow presents traffic
changes at different times of the day, while for the daily sampled series, these fine-grained variations
disappear but fluctuations associated with holidays emerge. On the other hand, the trend of macro-
economics dominates the yearly averaged patterns. These observations naturally call for a multiscale
analysis paradigm to disentangle complex temporal variations, where fine and coarse scales can
reflect the micro- and the macro-scopic information respectively. Especially for the time series
forecasting task, it is also notable that the future variation is jointly determined by the variations in
multiple scales. Therefore, in this paper, we attempt to design the forecasting model from a novel
view of multiscale-mixing, which is able to take advantage of both disentangled variations and
complementary forecasting capabilities from multiscale series simultaneously.
Technically, we propose TimeMixer with a multiscale mixing architecture that is able to extract
essential information from past variations by Past-Decomposable-Mixing (PDM) blocks and then
predicts the future series by the Future-Multipredictor-Mixing (FMM) block. Concretely, TimeMixer
first generates multiscale observations through average downsampling. Next, PDM adopts a decom-
posable design to better cope with distinct properties of seasonal and trend variations, by mixing
decomposed multiscale seasonal and trend components in fine-to-coarse and coarse-to-fine directions
separately. With our novel design, PDM is able to successfully aggregate the detailed seasonal
information starting from the finest series and dive into macroscopic trend components along with
the knowledge from coarser scales. In the forecasting phase, FMM ensembles multiple predictors to
utilize complementary forecasting capabilities from multiscale observations. With our meticulous
architecture, TimeMixer achieves the consistent state-of-the-art performances in both long-term and
short-term forecasting tasks with superior efficiency across all of our experiments, covering extensive
well-established benchmarks. Our contributions are summarized as follows:
• Going beyond previous methods, we tackle intricate temporal variations in series forecasting
from a novel view of multiscale mixing, taking advantage of disentangled variations and
complementary forecasting capabilities from multiscale series simultaneously.
• We propose TimeMixer as a simple but effective forecasting model, which enables the
combination of the multiscale information in both history extraction and future prediction
phases, empowered by our tailored decomposable and multiple-predictor mixing technique.
• TimeMixer achieves consistent state-of-the-art in performances in both long-term and short-
term forecasting tasks with superior efficiency on a wide range of benchmarks.

2 R ELATED W ORK
2.1 T EMPORAL M ODELING IN D EEP T IME S ERIES F ORECASTING

As the key problem in time series analysis (Wu et al., 2023a), temporal modeling has been widely
explored. According to foundation backbones, deep models can be roughly categorized into the
following four paradigms: RNN-, CNN-, Transformer- and MLP-based methods. Typically, CNN-
based models employ the convolution kernels along the time dimension to capture temporal patterns
(Wang et al., 2023; Hewage et al., 2020). And RNN-based methods adopt the recurrent structure
to model the temporal state transition (Lai et al., 2018; Zhao et al., 2017). However, both RNN-
and CNN-based methods suffer from the limited receptive field, limiting the long-term forecasting
capability. Recently, benefiting from the global modeling capacity, Transformer-based models have
been widely-acknowledged in long-term series forecasting (Zhou et al., 2021; Wu et al., 2021;
Liu et al., 2022b; Kitaev et al., 2020; Nie et al., 2023), which can capture the long-term temporal
dependencies adaptively with attention mechanism. Furthermore, multiple layer projection (MLP)
is also introduced to time series forecasting (Oreshkin et al., 2019; Challu et al., 2023; Zeng et al.,
2023), which achieves favourable performance in both forecasting performance and efficiency.
Additionally, several specific designs are proposed to better capture intricate temporal patterns,
including series decomposition and multi-periodicity analysis. Firstly, for the series decomposition,

2
Published as a conference paper at ICLR 2024

Autoformer (Wu et al., 2021) presents the series decomposition block based on moving average to
decompose complex temporal variations into seasonal and trend components. Afterwards, FEDformer
(Zhou et al., 2022b) enhances the series decomposition block with multiple kernels moving average.
DLinear (Zeng et al., 2023) utilizes the series decomposition as the pre-processing before linear
regression. MICN (Wang et al., 2023) also decomposes input series into seasonal and trend terms,
and then integrates the global and local context for forecasting. As for the multi-periodicity analysis,
N-BEATS (Oreshkin et al., 2019) fits the time series with multiple trigonometric basis functions.
FiLM (Zhou et al., 2022a) projects time series into Legendre Polynomials space, where different basis
functions correspond to different period components in the original series. Recently, TimesNet (Wu
et al., 2023a) adopts Fourier Transform to map time series into multiple components with different
period lengths and presents a modular architecture to process decomposed components.
Unlike the designs mentioned above, this paper explores the multiscale mixing architecture in time
series forecasting. Although there exist some models with temporal multiscale designs, such as
Pyraformer (Liu et al., 2021) with pyramidal attention and SCINet (Liu et al., 2022a) with a bifurcate
downsampling tree, their future predictions do not make use of the information at different scales
extracted from the past observations simultaneously. In TimeMixer, we present a new multiscale
mixing architecture with Past-Decomposable-Mixing to utilize the disentangled series for multiscale
representation learning and Future-Multipredictor-Mixing to ensemble the complementary forecasting
skills of multiscale series for better prediction.

2.2 M IXING N ETWORKS

Mixing is an effective way of information integration and has been applied to computer vision and
natural language processing. For instance, MLP-Mixer (Tolstikhin et al., 2021) designs a two-stage
mixing structure for image recognition, which mixes the channel information and patch information
successively with linear layers. FNet (Lee-Thorp et al., 2022) replaces attention layers in Transformer
with simple Fourier Transform, achieving the efficient token mixing of a sentence. In this paper, we
further explore the mixing structure in time series forecasting. Unlike previous designs, TimeMixer
presents a decomposable multi-scale mixing architecture and distinguishes the mixing methods in
both past information extraction and future prediction phases.

3 T IME M IXER
Given a series x with one or multiple observed variates, the main objective of time series forecasting
is to utilize past observations (length-P ) to obtain the most probable future prediction (length-F ). As
mentioned above, the key challenge of accurate forecasting is to tackle intricate temporal variations. In
this paper, we propose TimeMixer of multiscale-mixing, benefiting from disentangled variations and
complementary forecasting capabilities from multiscale series. Technically, TimeMixer consists of a
multiscale mixing architecture with Past-Decomposable-Mixing and Future-Multipredictor-Mixing
for past information extraction and future prediction respectively.

3.1 M ULTISCALE M IXING A RCHITECTURE

Time series of different scales naturally exhibit distinct properties, where fine scales mainly depict
detailed patterns and coarse scales highlight macroscopic variations (Mozer, 1991). This multiscale
view can inherently disentangle intricate variations in multiple components, thereby benefiting
temporal variation modeling. It is also notable that, especially for the forecasting task, multiscale time
series present different forecasting capabilities, due to their distinct dominating temporal patterns
(Ferreira et al., 2006). Therefore, we present TimeMixer in a multiscale mixing architecture to utilize
multiscale series with distinguishing designs for past extraction and future prediction phases.
As shown in Figure 1, to disentangle complex variations, we first downsample the past observations
x ∈ RP ×C into M scales by average pooling and finally obtain a set of multiscale time series
P
X = {x0 , · · · , xM }, where xm ∈ R⌊ 2m ⌋×C , m ∈ {0, · · · , M }, C denotes the variate number. The
lowest level series x0 = x is the input series, which contains the finest temporal variations, while the
highest-level series xM is for the macroscopic variations. Then we project these multiscale series
into deep features X 0 by the embedding layer, which can be formalized as X 0 = Embed(X). With
the above designs, we obtain the multiscale representations of input series.

3
Published as a conference paper at ICLR 2024

(a) Multiscale Time Series (b) Past Decomposable Mixing (c) Future Multipredictor Mixing

Lx Seasonal Mixing

Bottom-up + Predictor

Decomposition

Feed Forward
+ Predictor
Down-
Sampling
Top-down + Predictor

Prediction
+ Predictor +

Input Series Trend Mixing

Figure 1: Overall architecture of TimeMixer, which consists of Past-Decomposable Mixing and

Future-Multipredictor-Mixing for past observations and future predictions respectively.

Next, we utilize stacked Past-Decomposable-Mixing (PDM) blocks to mix past information across
different scales. For the l-th layer, the input is X l−1 and the process of PDM can be formalized as:
X l = PDM(X l−1 ), l ∈ {0, · · · , L}, (1)
P
where L is the total layer and X l = {xl0 , · · · , xlM }, xlm ∈ R⌊ 2m ⌋×dmodel denotes the mixed past
representations with dmodel channels. More details of PDM are described in the next section.
As for the future prediction phase, we adopt the Future-Multipredictor-Mixing (FMM) block to
ensemble extracted multiscale past information X L and generate future predictions, which is:
b = FMM(X L ),
x (2)
b ∈ RF ×C represents the final prediction. With the above designs, TimeMixer can successfully
where x
capture essential past information from disentangled multiscale observations and predict the future
with benefits from multiscale past information.

3.2 PAST D ECOMPOSABLE M IXING

We observe that for past observations, due to the complex nature of real-world series, even the coarsest
scale series present mixed variations. As shown in Figure 1, the series in the top layer still present
clear seasonality and trend simultaneously. It is notable that the seasonal and trend components hold
distinct properties in time series analysis (Cleveland et al., 1990), which corresponds to short-term
and long-term variations or stationary and non-stationary dynamics respectively. Therefore, instead
of directly mixing multiscale series as a whole, we propose the Past-Decomposable-Mixing (PDM)
block to mix the decomposed seasonal and trend components in multiple scales separately.
Concretely, for the l-th PDM block, we first decompose the multiscale time series Xl into seasonal
parts S l = {sl0 , · · · , slM } and trend parts Tl = {tl0 , · · · , tlM } by series decomposition block from
Autoformer (Wu et al., 2021). As the above analyzed, taking the distinct properties of seasonal-trend
parts into account, we apply the mixing operation to seasonal and trend terms separately to interact
information from multiple scales. Overall, the l-th PDM block can be formalized as:
slm , tlm = SeriesDecomp(xlm ), m ∈ {0, · · · , M },
(3)
 
X l = X l−1 + FeedForward S-Mix {slm }M l M



m=0 + T-Mix {t }
m m=0 ,

where FeedForward(·) contains two linear layers with intermediate GELU activation function for
information interaction among channels, S-Mix(·), T-Mix(·) denote seasonal and trend mixing.

Seasonal Mixing In seasonality analysis (Box & Jenkins, 1970), larger periods can be seen as
the aggregation of smaller periods, such as the weekly period of traffic flow formed by seven daily
changes, addressing the importance of detailed information in predicting future seasonal variations.
Therefore, in seasonal mixing, we adopt the bottom-up approach to incorporate information from
the lower-level fine-scale time series upwards, which can supplement detailed information to the
seasonality modeling of coarser scales. Technically, for the set of multiscale seasonal parts S l =

4
Published as a conference paper at ICLR 2024

g g
in in
ix ix
-M …
Up n-M
m- D ow
tt o p-
Bo … To t or
dic
… Pre

(a) Seasonal Mixing (b) Trend Mixing (c) Future Prediction

Figure 2: The temporal linear layer in seasonal mixing (a), trend mixing (b) and future prediction (c).

{sl0 , · · · , slM }, we use the Bottom-Up-Mixing layer for the m-th scale in a residual way to achieve
bottom-up seasonal information interaction, which can be formalized as:

for m: 1 → M do: slm = slm + Bottom-Up-Mixing(slm−1 ). (4)

where Bottom-Up-Mixing(·) is instantiated as two linear layers with an intermediate GELU activa-
P
tion function along the temporal dimension, whose input dimension is ⌊ 2m−1 ⌋ and output dimension
P
is ⌊ 2m ⌋. See Figure 2 for an intuitive understanding.
Trend Mixing Contrary to seasonal parts, for trend items, the detailed variations can introduce
noise in capturing macroscopic trend. Note that the upper coarse scale time series can easily provide
clear macro information than the lower level. Therefore, we adopt a top-down mixing method to
utilize the macro knowledge from coarser scales to guide the trend modeling of finer scales.
Technically, for multiscale trend components Tl = {tl0 , · · · , tlM }, we adopt the Top-Down-Mixing
layer for the m-th scale in a residual way to achieve top-down trend information interaction:

for m: (M − 1) → 0 do: tlm = tlm + Top-Down-Mixing(tlm+1 ), (5)

where Top-Down-Mixing(·) is two linear layers with an intermediate GELU activation function,
P
whose input dimension is ⌊ 2m+1 ⌋ and output dimension is ⌊ 2Pm ⌋ as shown in Figure 2.
Empowered by seasonal and trend mixing, PDM progressively aggregates the detailed seasonal
information from fine to coarse and dive into the macroscopic trend information with prior knowledge
from coarser scales, eventually achieving the multiscale mixing in past information extraction.

3.3 F UTURE M ULTIPREDICTOR M IXING

After L PDM blocks, we obtain the multiscale past information as X L = {xL L L

0 , · · · , xM }, xm ∈
P
R⌊ 2m ⌋×dmodel . Since the series in different scales presents different dominating variations, their
predictions also present different capabilities. To fully utilize the multiscale information, we propose
to aggregate predictions from multiscale series and present Future-Multipredictor-Mixing block as:
M
X
bm = Predictorm (xL
x m ), m ∈ {0, · · · , M }, x
b= x
bm , (6)
m=0

where xbm ∈ RF ×C represents the future prediction from the m-th scale series and the final output is
b ∈ RF ×C . Predictorm (·) denotes the predictor of the m-th scale series, which firstly adopts one
x
single linear layer to directly regress length-F future from length-⌊ 2Pm ⌋ extracted past information
(Figure 2) and then projects deep representations into C variates. Note that FMM is an ensemble of
multiple predictors, where different predictors are based on past information from different scales,
enabling FMM to integrate complementary forecasting capabilities of mixed multiscale series.

4 E XPERIMENTS
We conduct extensive experiments to evaluate the performance and efficiency of TimeMixer, covering
long-term and short-term forecasting, including 18 real-world benchmarks and 15 baselines. The
detailed model and experiment configurations are summarized in Appendix A.

5
Published as a conference paper at ICLR 2024

Table 1: Summary of benchmarks. Forecastability is one minus the entropy of Fourier domain.
Tasks Dataset Variate Predict Length Frequency Forecastability Information
ETT (4 subsets) 7 96∼720 15 mins 0.46 Temperature
Long-term Weather 21 96∼720 10 mins 0.75 Weather
forecasting Solar-Energy 137 96∼720 10min 0.33 Electricity
Electricity 321 96∼720 Hourly 0.77 Electricity
Traffic 862 96∼720 Hourly 0.68 Transportation
Short-term PEMS (4 subsets) 170∼883 12 5min 0.55 Traffic network
forecasting M4 (6 subsets) 1 6∼48 Hourly∼Yearly 0.47 Database

Benchmarks For long-term forecasting, we experiment on 8 well-established benchmarks: ETT

datasets (including 4 subsets: ETTh1, ETTh2, ETTm1, ETTm2), Weather, Solar-Energy, Electricity,
and Traffic following (Zhou et al., 2021; Wu et al., 2021; Liu et al., 2022a). For short-term forecasting,
we adopt the PeMS (Chen et al., 2001) which contains four public traffic network datasets (PEMS03,
PEMS04, PEMS07, PEMS08), and M4 dataset which involves 100,000 different time series collected
in different frequencies. Furthermore, we measure the forecastability (Goerg, 2013) of all datasets.
It is observed that ETT, M4, and Solar-Energy exhibit relatively low forecastability, indicating the
challenges in these benchmarks. More information is summarized in Table 1.
Baselines We compare TimeMixer with 15 baselines, which comprise the state-of-the-art long-term
forecasting model PatchTST (2023) and advanced short-term forecasting models TimesNet (2023a)
and SCINet (2022a), as well as other competitive models including Crossformer (2023), MICN
(2023), FiLM (2022a), DLinear (2023), LightTS (2022) ,FEDformer (2022b), Stationary (2022b),
Pyraformer (2021), Autoformer (2021), Informer (2021), N-HiTS (2023) and N-BEATS (2019).
Unified experiment settings Note that experimental results reported by the above mentioned
baselines cannot be compared directly due to different choices of input length and hyper-parameter
searching strategy. For fairness, we make a great effort to provide two types of experiments. In the
main text, we align the input length of all baselines and report results averaged from three repeats (see
Appendix C for error bars). In Appendix, to compare the upper bound of models, we also conduct a
comprehensive hyperparameter searching and report the best results in Table 14 of Appendix.
Implementation details All the experiments are implemented in PyTorch (Paszke et al., 2019) and
conducted on a single NVIDIA A100 80GB GPU. We utilize the L2 loss for model training. The
number of scales M is set according to the time series length to trade off performance and efficiency.

4.1 M AIN R ESULTS

Long-term forecasting As shown in Table 2, TimeMixer achieves consistent state-of-the-art

performance in all benchmarks, covering a large variety of series with different frequencies, variate
numbers and real-world scenarios. Especially, TimeMixer outperforms PatchTST by a considerable
margin, with a 9.4% MSE reduction in Weather and a 24.7% MSE reduction in Solar-Energy. It is
worth noting that TimeMixer exhibits good performance even for datasets with low forecastability,
such as Solar-Energy and ETT, further proving the generality and effectiveness of TimeMixer.
Table 2: Long-term forecasting results. All the results are averaged from 4 different prediction
lengths, that is {96, 192, 336, 720}. A lower MSE or MAE indicates a better prediction. We fix the
input length as 96 for all experiments. See Table 13 in Appendix for the full results.
TimeMixer PatchTST TimesNet Crossformer MICN FiLM DLinear FEDformer Stationary Autoformer Informer
Models
(Ours) (2023) (2023a) (2023) (2023) (2022a) (2023) (2022b) (2022b) (2021) (2021)
Metric MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE
Weather 0.240 0.271 0.265 0.285 0.251 0.294 0.264 0.320 0.268 0.321 0.271 0.291 0.265 0.315 0.309 0.360 0.288 0.314 0.338 0.382 0.634 0.548
Solar-Energy 0.216 0.280 0.287 0.333 0.403 0.374 0.406 0.442 0.283 0.358 0.380 0.371 0.330 0.401 0.328 0.383 0.350 0.390 0.586 0.557 0.331 0.381
Electricity 0.182 0.272 0.216 0.318 0.193 0.304 0.244 0.334 0.196 0.309 0.223 0.302 0.225 0.319 0.214 0.327 0.193 0.296 0.227 0.338 0.311 0.397
Traffic 0.484 0.297 0.529 0.341 0.620 0.336 0.667 0.426 0.593 0.356 0.637 0.384 0.625 0.383 0.610 0.376 0.624 0.340 0.628 0.379 0.764 0.416
ETTh1 0.447 0.440 0.516 0.484 0.495 0.450 0.529 0.522 0.475 0.480 0.516 0.483 0.461 0.457 0.498 0.484 0.570 0.537 0.496 0.487 1.040 0.795
ETTh2 0.364 0.395 0.391 0.411 0.414 0.427 0.942 0.684 0.574 0.531 0.402 0.420 0.563 0.519 0.437 0.449 0.526 0.516 0.450 0.459 4.431 1.729
ETTm1 0.381 0.395 0.406 0.407 0.400 0.406 0.513 0.495 0.423 0.422 0.411 0.402 0.404 0.408 0.448 0.452 0.481 0.456 0.588 0.517 0.961 0.734
ETTm2 0.275 0.323 0.290 0.334 0.291 0.333 0.757 0.610 0.353 0.402 0.287 0.329 0.354 0.402 0.305 0.349 0.306 0.347 0.327 0.371 1.410 0.810

6
Published as a conference paper at ICLR 2024

Table 3: Short-term forecasting results in the PEMS datasets with multiple variates. All input lengths
are 96 and prediction lengths are 12. A lower MAE, MAPE or RMSE indicates a better prediction.
TimeMixer SCINet Crossformer PatchTST TimesNet MICN FiLM DLinear FEDformer Stationary Autoformer Informer
Models
(Ours) (2022a) (2023) (2023) (2023a) (2023) (2022a) (2023) (2022b) (2022b) (2021) (2021)
MAE 14.63 15.97 15.64 18.95 16.41 15.71 21.36 19.70 19.00 17.64 18.08 19.19
PEMS03 MAPE 14.54 15.89 15.74 17.29 15.17 15.67 18.35 18.35 18.57 17.56 18.75 19.58
RMSE 23.28 25.20 25.56 30.15 26.72 24.55 35.07 32.35 30.05 28.37 27.82 32.70
MAE 19.21 20.35 20.38 24.86 21.63 21.62 26.74 24.62 26.51 22.34 25.00 22.05
PEMS04 MAPE 12.53 12.84 12.84 16.65 13.15 13.53 16.46 16.12 16.76 14.85 16.70 14.88
RMSE 30.92 32.31 32.41 40.46 34.90 34.39 42.86 39.51 41.81 35.47 38.02 36.20
MAE 20.57 22.79 22.54 27.87 25.12 22.28 28.76 28.65 27.92 26.02 26.92 27.26
PEMS07 MAPE 8.62 9.41 9.38 12.69 10.60 9.57 11.21 12.15 12.29 11.75 11.83 11.63
RMSE 33.59 35.61 35.49 42.56 40.71 35.40 45.85 45.02 42.29 42.34 40.60 45.81
MAE 15.22 17.38 17.56 20.35 19.01 17.76 22.11 20.26 20.56 19.29 20.47 20.96
PEMS08 MAPE 9.67 10.80 10.92 13.15 11.83 10.76 12.81 12.09 12.41 12.21 12.27 13.20
RMSE 24.26 27.34 27.21 31.04 30.65 27.26 35.13 32.38 32.97 38.62 31.52 30.61

Table 4: Short-term forecasting results in the M4 dataset with a single variate. All prediction lengths
are in [6, 48]. A lower SMAPE, MASE or OWA indicates a better prediction. ∗. in the Transformers
indicates the name of ∗former. Stationary means the Non-stationary Transformer.
TimeMixer TimesNet N-HiTS N-BEATS∗ SCINet PatchTST MICN FiLM LightTS DLinear FED. Stationary Auto. Pyra. In.
Models
(Ours) (2023a) (2023) (2019) (2022a) (2023) (2023) (2022a) (2022) (2023) (2022b) (2022b) (2021) (2021) (2021)
SMAPE 13.206 13.387 13.418 13.436 18.605 16.463 25.022 17.431 14.247 16.965 13.728 13.717 13.974 15.530 14.727
Yearly

MASE 2.916 2.996 3.045 3.043 4.471 3.967 7.162 4.043 3.109 4.283 3.048 3.078 3.134 3.711 3.418
OWA 0.776 0.786 0.793 0.794 1.132 1.003 1.667 1.042 0.827 1.058 0.803 0.807 0.822 0.942 0.881
Others Monthly Quarterly

SMAPE 9.996 10.100 10.202 10.124 14.871 10.644 15.214 12.925 11.364 12.145 10.792 10.958 11.338 15.449 11.360
MASE 1.166 1.182 1.194 1.169 2.054 1.278 1.963 1.664 1.328 1.520 1.283 1.325 1.365 2.350 1.401
OWA 0.825 0.890 0.899 0.886 1.424 0.949 1.407 1.193 1.000 1.106 0.958 0.981 1.012 1.558 1.027
SMAPE 12.605 12.670 12.791 12.677 14.925 13.399 16.943 15.407 14.014 13.514 14.260 13.917 13.958 17.642 14.062
MASE 0.919 0.933 0.969 0.937 1.131 1.031 1.442 1.298 1.053 1.037 1.102 1.097 1.103 1.913 1.141
OWA 0.869 0.878 0.899 0.880 1.027 0.949 1.265 1.144 0.981 0.956 1.012 0.998 1.002 1.511 1.024
SMAPE 4.564 4.891 5.061 4.925 16.655 6.558 41.985 7.134 15.880 6.709 4.954 6.302 5.485 24.786 24.460
MASE 3.115 3.302 3.216 3.391 15.034 4.511 62.734 5.09 11.434 4.953 3.264 4.064 3.865 18.581 20.960
OWA 0.982 1.035 1.040 1.053 4.123 1.401 14.313 1.553 3.474 1.487 1.036 1.304 1.187 5.538 5.879
Weighted
Average

SMAPE 11.723 11.829 11.927 11.851 15.542 13.152 19.638 14.863 13.525 13.639 12.840 12.780 12.909 16.987 14.086
MASE 1.559 1.585 1.613 1.559 2.816 1.945 5.947 2.207 2.111 2.095 1.701 1.756 1.771 3.265 2.718
OWA 0.840 0.851 0.861 0.855 1.309 0.998 2.279 1.125 1.051 1.051 0.918 0.930 0.939 1.480 1.230
∗ The original paper of N-BEATS (2019) adopts a special ensemble method to promote the performance.
For fair comparisons, we remove the ensemble and only compare the pure forecasting models.

Short-term forecasting TimeMixer also shows great performance in short-term forecasting under
both multivariate and univariate settings (Table 3-4). For PeMS benchmarks that record multiple
time series of citywide traffic networks, due to the complex spatiotemporal correlations among
multiple variates, many advanced models degenerate a lot in this task, such as PatchTST (2023)
and DLinear (2023), which adopt the channel independence design. In contrast, TimeMixer still
performs favourablely in this challenging problem, verifying its effectiveness in handling complex
multivariate time series forecasting. As for the M4 dataset for univariate forecasting, it contains
various temporal variations under different sampling frequencies, including hourly, daily, weekly,
monthly, quarterly, and yearly, which exhibits low predictability and distinctive characteristics across
different frequencies. Remarkably, Timemixer consistently performs best across all frequencies,
affirming the multiscale mixing architecture’s capacity in modeling complex temporal variations.

4.2 M ODEL A NALYSIS

Ablations To verify the effectiveness of each component of TimeMixer, we provide detailed ablation
study on every possible design in both Past-Decomposable-Mixing and Future-Multipredictor-Mixing
blocks on all 18 experiment benchmarks. From Table 5, we have the following observations.
The exclusion of Future-Multipredictor-Mixing in ablation ② results in a significant decrease in the
model’s forecasting accuracy for both short and long-term predictions. This demonstrates that mixing
future predictions from multiscale series can effectively boost the model performance.
For the past mixing, we verify the effectiveness by removing or replacing components gradually. In
ablations ③ and ④ that remove seasonal mixing and trend mixing respectively, also cause a decline

7
Published as a conference paper at ICLR 2024

Table 5: Ablations on both PDM (Decompose, Season Mixing, Trend Mixing) and FMM blocks in M4,
PEMS04 and predict-336 setting of ETTm1. ↗ indicates the bottom-up mixing while ↙ indicates
top-down. A check mark ✓ and a wrong mark × indicate with and without certain components
respectively. ① is the official design in TimeMixer (See Appendix F for complete ablation results).
Past mixing Future mixing M4 PEMS04 ETTm1
Case Decompose
Seasonal Trend Multipredictor SMAPE MASE OWA MAE MAPE RMSE MSE MAE
① ✓ ↗ ↙ ✓ 11.723 1.559 0.840 19.21 12.53 30.92 0.390 0.404
② ✓ ↗ ↙ × 12.503 1.634 0.925 21.67 13.45 34.89 0.402 0.415
③ ✓ × ↙ ✓ 13.051 1.676 0.962 24.49 16.28 38.79 0.411 0.427
④ ✓ ↗ × ✓ 12.911 1.655 0.941 22.91 15.02 37.04 0.405 0.414
⑤ ✓ ↙ ↙ ✓ 12.008 1.628 0.871 20.78 13.02 32.47 0.392 0.413
⑥ ✓ ↗ ↗ ✓ 11.978 1.626 0.859 21.09 13.78 33.11 0.396 0.415
⑦ ✓ ↙ ↗ ✓ 13.012 1.657 0.954 22.27 15.14 34.67 0.412 0.429
⑧ × ↗ ✓ 11.975 1.617 0.851 21.51 13.47 34.81 0.395 0.408
⑨ × ↙ ✓ 11.973 1.622 0.850 21.79 14.03 35.23 0.393 0.406
⑩ × × ✓ 12.468 1.671 0.916 24.87 16.66 39.48 0.405 0.412

of performance. This illustrates that solely relying on seasonal or trend information interaction is
insufficient for accurate predictions. Furthermore, in both ablations ⑤ and ⑥, we employed the same
mixing approach for both seasons and trends. However, it cannot bring better predictive performance.
Similar situation occurs in ⑦ that adopts opposite mixing strategies to our design. These results
demonstrate the effectiveness of our design in both bottom-up seasonal mixing and top-down trend
mixing. Concurrently, in ablations ⑧ and ⑨, we opted to eliminate the decomposition architecture and
mix the multiscale series directly. However, without decomposition, neither bottom-up nor top-down
mixing method can achieve a good performance, indicating the necessity of season-trend separate
mixing. Furthermore, in ablations ⑩, eliminating the entire Past-Decomposable-Mixing block causes
a serious drop in the model’s predictive performance. The above findings highlight the substantial
influence of an appropriate past mixing method on the final performance of the model. Starting from
the insights in time series, TimeMixer presents the best mixing method in past information extraction.
Seasonal and trend mixing visualization To provide an intuitive understanding of PDM, we
visualize temporal linear weights for seasonal mixing and trend mixing in Figure 3(a)∼(b). We find
that the seasonal and trend items present distinct mixing properties, where the seasonal mixing layer
presents periodic changes (repeated blue lines in (a)) and the trend mixing layer is dominated by
local aggregations (the dominating diagonal yellow line in (b)). This also verifies the necessity of
adopting separate mixing techniques for seasonal and trend terms. Furthermore, Figure 3(c) shows
the predictions of season and trend terms in fine (scale 0) and coarse (scale 3) scales. We can observe
that the seasonal terms of fine-scale and trend parts of coarse-scale are crucial for accurate predictions.
This observation provides insights for our design in utilizing bottom-up mixing for seasonal terms
and top-down mixing for trend components.
-0.05 -0.00 -0.50 -0.10 -0.15
0
Mixing

0 -0.10
24
12 -0.05

0.00
Season

48
24
0.05
36 72
0.10
Trend

48 96 0.15
0 24 48 72 96 0 24 48
Scale0 Scale3

(a) Seasonal Mixing Weights (b) Trend Mixing Weights (c) Multiscale Season-trend Predictions
(bottom-up: from 96 to 48) (top-down: from 48 to 96) (input-96-predict-96)

Figure 3: Visualization of temporal linear weights in seasonal mixing (Eq. 4), trend mixing (Eq. 5),
and predictions from multiscale season-trend items. All the experiments are on the ETTh1 dataset
under the input-96-predict-96 setting.

8
Published as a conference paper at ICLR 2024

Input-96-predict-96

(a) Multiscale mixing (b) Scale 0 (c) Scale 1 (d) Scale 2 (e) Scale 3

xL
Figure 4: Visualization of predictions from different scales (b m in Eq. 6) on the input-96-predict-96
settings of the ETTh1 dataset. The implementation details are included in Appendix A.

(a) Memory Efficiency Analysis (b) Running Time Efficiency Analysis

Figure 5: Efficiency analysis in both GPU memory and running time. The results are recorded on the
ETTh1 dataset with batch size as 16. The running time is averaged from 102 iterations.

Multipredictor visualization To provide an intuitive understanding of the forecasting skills of

multiscale series, we plot the forecasting results from different scales for qualitative comparison.
Figure 4(a) presents the overall prediction of our model with Future-Multipredictor-Mixing, which
indicates accurate prediction according to the future variations using mixed scales. To study the
component of each individual scale, we demonstrate the prediction results for each scale in Figure
4(b)∼(e). Specifically, prediction results from fine-scale time series concentrate more on the detailed
variations of time series and capture seasonal patterns with greater precision. In contrast, as shown in
Figure 4(c)∼(e), with multiple downsampling, the predictions from coarse-scale series focus more
on macro trends. The above results also highlight the benefits of Future-Multipredictor-Mixing in
utilizing complementary forecasting skills from multiscale series.
0.60
Analysis on Number of Scales
Efficiency analysis We compare the running memory and predict-720
predict-336
time against the latest state-of-the-art models in Figure 5 un- predict-192
0.52
der the training phase, where TimeMixer consistently demon- predict-96
strates favorable efficiency, in terms of both GPU memory
MSE

and running time, for various series lengths (ranging from 192
to 3072), in addition to the consistent state-of-the-art perfor-
mances for both long-term and short-term forecasting tasks.
0.34

Analysis on number of scales We explore the impact from 0.30

1 2 3 4
Number of scales
the number of scales (M ) in Figure 6 under different series
lengths. Specifically, when M increases, the performance gain Figure 6: Analysis on number of scales
declines for shorter prediction lengths. In contrast, for longer on ETTm1 dataset.
prediction lengths, the performance improves more as M increases. Therefore, we set M as 3 for
long-term forecast and 1 for short-term forecast to trade off performance and efficiency.

5 C ONCLUSION
We presented TimeMixer with a multiscale mixing architecture to tackle the intricate temporal
variations in time series forecasting. Empowered by Past-Decomposable-Mixing and Future-
Multipredictor-Mixing blocks, TimeMixer took advantage of both disentangled variations and
complementary forecasting capabilities. In all of our experiments, TimeMixer achieved consis-
tent state-of-the-art performances in both long-term and short-term forecasting tasks. Moreover,
benefiting from the fully MLP-based architecture, TimeMixer demonstrated favorable run-time
efficiency. Detailed visualizations and ablations are included to provide insights for our design.

9
Published as a conference paper at ICLR 2024

6 E THICS S TATEMENT

Our work only focuses on the scientific problem, so there is no potential ethical risk.

7 R EPRODUCIBILITY S TATEMENT

We involve the implementation details in Appendix A, including dataset descriptions, metric calcu-
lation and experiment configuration. The source code is provided in supplementary materials and
public in GitHub (https://github.com/kwuking/TimeMixer) for reproducibility.

ACKNOWLEDGMENTS

This work was supported by Ant Group through CCF-Ant Research Fund.

R EFERENCES
G. E. P. Box and Gwilym M. Jenkins. Time series analysis, forecasting and control. 1970.

Cristian Challu, Kin G Olivares, Boris N Oreshkin, Federico Garza, Max Mergenthaler, and Artur
Dubrawski. N-hits: Neural hierarchical interpolation for time series forecasting. AAAI, 2023.

Chao Chen, Karl F. Petty, Alexander Skabardonis, Pravin Pratap Varaiya, and Zhanfeng Jia. Freeway
performance measurement system: Mining loop detector data. Transportation Research Record,
2001.

Si-An Chen, Chun-Liang Li, Nate Yoder, Sercan O. Arik, and Tomas Pfister. Tsmixer: An all-mlp
architecture for time series forecasting. arXiv preprint arXiv:2303.06053, 2023.

Robert B Cleveland, William S Cleveland, Jean E McRae, and Irma Terpenning. STL: A seasonal-
trend decomposition. Journal of Official Statistics, 1990.

Marco A. R. Ferreira, Michael A. West, Herbert K. H. Lee, and David M. Higdon. Multi-scale and
hidden resolution time series models. Bayesian Analysis, 2006.

Georg Goerg. Forecastable component analysis. ICML, 2013.

Clive William John Granger and Paul Newbold. Forecasting economic time series. Academic press,
2014.

Pradeep Hewage, Ardhendu Behera, Marcello Trovati, Ella Pereira, Morteza Ghahremani, Francesco
Palmieri, and Yonghuai Liu. Temporal convolutional neural (TCN) network for an effective weather
forecasting using time-series data from the local weather station. Soft Computing, 2020.

Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization. ICLR, 2015.

Nikita Kitaev, Lukasz Kaiser, and Anselm Levskaya. Reformer: The efficient transformer. ICLR,
2020.

Guokun Lai, Wei-Cheng Chang, Yiming Yang, and Hanxiao Liu. Modeling long-and short-term
temporal patterns with deep neural networks. SIGIR, 2018.

James Lee-Thorp, Joshua Ainslie, Ilya Eckstein, and Santiago Ontanon. Fnet: Mixing tokens with
fourier transforms. NAACL, 2022.

Zhe Li, Zhongwen Rao, Lujia Pan, and Zenglin Xu. Mts-mixers: Multivariate time series forecasting
via factorized temporal and channel mixing. arXiv preprint arXiv:2302.04501, 2023.

Minhao Liu, Ailing Zeng, Muxi Chen, Zhijian Xu, Qiuxia Lai, Lingna Ma, and Qiang Xu. SCINet:
time series modeling and forecasting with sample convolution and interaction. NeurIPS, 2022a.

10
Published as a conference paper at ICLR 2024

Shizhan Liu, Hang Yu, Cong Liao, Jianguo Li, Weiyao Lin, Alex X Liu, and Schahram Dust-
dar. Pyraformer: Low-complexity pyramidal attention for long-range time series modeling and
forecasting. ICLR, 2021.
Yong Liu, Haixu Wu, Jianmin Wang, and Mingsheng Long. Non-stationary transformers: Rethinking
the stationarity in time series forecasting. NeurIPS, 2022b.
Luis Martín, Luis F Zarzalejo, Jesus Polo, Ana Navarro, Ruth Marchante, and Marco Cony. Prediction
of global solar irradiance based on time series analysis: Application to solar thermal power plants
energy production planning. Solar Energy, 2010.
Michael C. Mozer. Induction of multiscale temporal structure. NeurIPS, 1991.
Yuqi Nie, Nam H Nguyen, Phanwadee Sinthong, and Jayant Kalagnanam. A time series is worth 64
words: Long-term forecasting with transformers. ICLR, 2023.
Boris N Oreshkin, Dmitri Carpov, Nicolas Chapados, and Yoshua Bengio. N-BEATS: Neural basis
expansion analysis for interpretable time series forecasting. ICLR, 2019.
Adam Paszke, S. Gross, Francisco Massa, A. Lerer, James Bradbury, Gregory Chanan, Trevor Killeen,
Z. Lin, N. Gimelshein, L. Antiga, Alban Desmaison, Andreas Köpf, Edward Yang, Zach DeVito,
Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu Fang, Junjie Bai, and
Soumith Chintala. Pytorch: An imperative style, high-performance deep learning library. NeurIPS,
2019.
Zheng Qian, Yan Pei, Hamidreza Zareipour, and Niya Chen. A review and discussion of
decomposition-based hybrid models for wind energy forecasting applications. Applied energy,
2019.
Yao Qin, Dongjin Song, Haifeng Chen, Wei Cheng, Guofei Jiang, and Garrison Cottrell. A dual-stage
attention-based recurrent neural network for time series prediction. IJCAI, 2017.
David Salinas, Valentin Flunkert, Jan Gasthaus, and Tim Januschowski. DeepAR: Probabilistic
forecasting with autoregressive recurrent networks. International Journal of Forecasting, 2020.
Amin Shabani, Amir Abdi, Lili Meng, and Tristan Sylvain. Scaleformer: iterative multi-scale refining
trans- formers for time series forecasting. ICLR, 2023.
Ilya Tolstikhin, Neil Houlsby, Alexander Kolesnikov, Lucas Beyer, Xiaohua Zhai, Thomas Un-
terthiner, Jessica Yung, Andreas Steiner, Daniel Keysers, Jakob Uszkoreit, Mario Lucic, and
Alexey Dosovitskiy. MLP-Mixer: An all-MLP architecture for vision. NeurIPS, 2021.
Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Lukasz
Kaiser, and Illia Polosukhin. Attention is all you need. NeurIPS, 2017.
Huiqiang Wang, Jian Peng, Feihu Huang, Jince Wang, Junhui Chen, and Yifei Xiao. MICN: Multi-
scale local and global context modeling for long-term series forecasting. ICLR, 2023.
Haixu Wu, Jiehui Xu, Jianmin Wang, and Mingsheng Long. Autoformer: Decomposition transformers
with Auto-Correlation for long-term series forecasting. NeurIPS, 2021.
Haixu Wu, Tengge Hu, Yong Liu, Hang Zhou, Jianmin Wang, and Mingsheng Long. TimesNet:
Temporal 2d-variation modeling for general time series analysis. ICLR, 2023a.
Haixu Wu, Hang Zhou, Mingsheng Long, and Jianmin Wang. Interpretable weather forecasting for
worldwide stations with a unified deep model. Nature Machine Intelligence, 2023b.
Xueyan Yin, Genze Wu, Jinze Wei, Yanming Shen, Heng Qi, and Baocai Yin. Deep learning on
traffic prediction: Methods, analysis, and future directions. IEEE Transactions on Intelligent
Transportation Systems, 2021.
Ailing Zeng, Muxi Chen, Lei Zhang, and Qiang Xu. Are transformers effective for time series
forecasting? AAAI, 2023.

11
Published as a conference paper at ICLR 2024

Tianping Zhang, Yizhuo Zhang, Wei Cao, Jiang Bian, Xiaohan Yi, Shun Zheng, and Jian Li. Less is
more: Fast multivariate time series forecasting with light sampling-oriented mlp structures. arXiv
preprint arXiv:2207.01186, 2022.

Yunhao Zhang and Junchi Yan. Crossformer: Transformer utilizing cross-dimension dependency for
multivariate time series forecasting. ICLR, 2023.

Zheng Zhao, Weihai Chen, Xingming Wu, Peter CY Chen, and Jingmeng Liu. Lstm network: a deep
learning approach for short-term traffic forecast. IET Intelligent Transport Systems, 2017.

Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, and Wancai Zhang.
Informer: Beyond efficient transformer for long sequence time-series forecasting. AAAI, 2021.

Tian Zhou, Ziqing Ma, Qingsong Wen, Liang Sun, Tao Yao, Wotao Yin, Rong Jin, et al. Film:
Frequency improved legendre memory model for long-term time series forecasting. NeurIPS,
2022a.

Tian Zhou, Ziqing Ma, Qingsong Wen, Xue Wang, Liang Sun, and Rong Jin. FEDformer: Frequency
enhanced decomposed transformer for long-term series forecasting. ICML, 2022b.

A I MPLEMENTATION D ETAILS

We summarized details of datasets, evaluation metrics, experiments and visualizations in this section.

Datasets details We evaluate the performance of different models for long-term forecasting on
8 well-established datasets, including Weather, Traffic, Electricity, Solar-Energy, and ETT datasets
(ETTh1, ETTh2, ETTm1, ETTm2). Furthermore, we adopt PeMS and M4 datasets for short-term
forecasting. We detail the descriptions of the dataset in Table 6.

Table 6: Dataset detailed descriptions. The dataset size is organized in (Train, Validation, Test).
Tasks Dataset Dim Series Length Dataset Size Frequency Forecastability∗ Information
ETTm1 7 {96, 192, 336, 720} (34465, 11521, 11521) 15min 0.46 Temperature

ETTm2 7 {96, 192, 336, 720} (34465, 11521, 11521) 15min 0.55 Temperature

ETTh1 7 {96, 192, 336, 720} (8545, 2881, 2881) 15 min 0.38 Temperature

ETTh2 7 {96, 192, 336, 720} (8545, 2881, 2881) 15 min 0.45 Temperature

Long-term Electricity 321 {96, 192, 336, 720} (18317, 2633, 5261) Hourly 0.77 Electricity

Forecasting Traffic 862 {96, 192, 336, 720} (12185, 1757, 3509) Hourly 0.68 Transportation

Weather 21 {96, 192, 336, 720} (36792, 5271, 10540) 10 min 0.75 Weather

Solar-Energy 137 {96, 192, 336, 720} (36601, 5161, 10417) 10min 0.33 Electricity

PEMS03 358 12 (15617,5135,5135) 5min 0.65 Transportation

PEMS04 307 12 (10172,3375,3375) 5min 0.45 Transportation

PEMS07 883 12 (16911,5622,5622) 5min 0.58 Transportation

Short-term PEMS08 170 12 (10690,3548,265) 5min 0.52 Transportation

Forecasting M4-Yearly 1 6 (23000, 0, 23000) Yearly 0.43 Demographic

M4-Quarterly 1 8 (24000, 0, 24000) Quarterly 0.47 Finance

M4-Monthly 1 18 (48000, 0, 48000) Monthly 0.44 Industry

M4-Weakly 1 13 (359, 0, 359) Weakly 0.43 Macro

M4-Daily 1 14 (4227, 0, 4227) Daily 0.44 Micro

M4-Hourly 1 48 (414, 0, 414) Hourly 0.46 Other

∗ The forecastability is calculated by one minus the entropy of Fourier decomposition of time series (Goerg, 2013). A
larger value indicates better predictability.

12
Published as a conference paper at ICLR 2024

Metric details Regarding metrics, we utilize the mean square error (MSE) and mean absolute
error (MAE) for long-term forecasting. In the case of short-term forecasting, we follow the metrics
of SCINet (Liu et al., 2022a) on the PeMS datasets, including mean absolute error (MAE), mean
absolute percentage error (MAPE), root mean squared error (RMSE). As for the M4 datasets, we
follow the methodology of N-BEATS (Oreshkin et al., 2019) and implement the symmetric mean
absolute percentage error (SMAPE), mean absolute scaled error (MASE), and overall weighted
average (OWA) as metrics. It is worth noting that OWA is a specific metric utilized in the M4
competition. The calculations of these metrics are:
F F
b i )2 ) 21 ,
X X
RMSE = ( (Xi − X MAE = |Xi − X
b i |,
i=1 i=1
F F
200 X |Xi − X b i| 100 X |Xi − Xb i|
SMAPE = , MAPE = ,
F i=1 |Xi | + |X
b i| F i=1 |Xi |
F  
1 X |Xi − X
b i| 1 SMAPE MASE
MASE = F
, OWA = + ,
F i=1 1 2 SMAPENaïve2 MASENaïve2
P
F −s j=s+1 |Xj − Xj−s |

b ∈ RF ×C are the ground truth and prediction results of

where s is the periodicity of the data. X, X
the future with F time pints and C dimensions. Xi means the i-th future time point.

Experiment details All experiments were run three times, implemented in Pytorch (Paszke et al.,
2019), and conducted on a single NVIDIA A100 80GB GPU. We set the initial learning rate as 10−2
or 10−3 and used the ADAM optimizer (Kingma & Ba, 2015) with L2 loss for model optimization.
And the batch size was set to be 8 between 128. By default, TimeMixer contains 2 Past Decomposable
Mixing blocks. We choose the number of scales M according to the length of the time series to achieve
a balance between performance and efficiency. To handle longer series in long-term forecasting, we
set M to 3. As for short-term forecasting with limited series length, we set M to 1. Detailed model
configuration information is presented in Table 7.

Visualization details To verify complementary forecasting capabilities of multiscale series, we

fix the PDM and train a new predictor for the feature at each scale with the ground truth future as
supervision in Figure 4; Figure 3(c) also utilizes the same operations. Especially for Figure 4, we also
bm , m ∈ {0, · · · , M }
provide the visualization of directly plotting the output of each predictor, i.e. x
b= M
P
in Eq. 6. Note that in FMM, we adopt the sum ensemble x m=0 mx
b as the final output, the scale
of each plotted cure is around M1+1 of ground truth, while we can still observe the distinct forecasting
capability of series in different scales. For clearness, we also plot the (M + 1) × x bm in the second
row of Figure 7, where the visualizations are similar to Figure 4.

(b) Scale 0 (c) Scale 1 (d) Scale 2 (e) Scale 3

(a) Multiscale mixing

(f) Scale 0 × (𝑀 + 1) (g) Scale 1 × (𝑀 + 1) (h) Scale 2 × (𝑀 + 1) (i) Scale 3 × (𝑀 + 1)

Figure 7: Direct visualization of predictions from different scales xL

(b
in Eq. 6) on the input-96-
m
predict-96 settings of the ETTh1 dataset. We multiply the (M + 1) by the predictions of each scale
in the second row.

13
Published as a conference paper at ICLR 2024

Table 7: Experiment configuration of TimeMixer. All the experiments use the ADAM (2015)
optimizer with the default hyperparameter configuration for (β1 , β2 ) as (0.9, 0.999).
Model Hyper-parameter Training Process
Dataset / Configurations
∗
M (Equ. 1) Layers dmodel LR Loss Batch Size Epochs
ETTh1 3 2 16 10−2 MSE 128 10
−2
ETTh2 3 2 16 10 MSE 128 10
−2
ETTm1 3 2 16 10 MSE 128 10
−2
ETTm2 3 2 32 10 MSE 128 10
−2
Weather 3 2 16 10 MSE 128 20
−2
Electricity 3 2 16 10 MSE 32 20
−2
Solar-Energy 3 2 128 10 MSE 32 20
−2
Traffic 3 2 32 10 MSE 8 20
−3
PEMS 1 5 128 10 MSE 32 10
−2
M4 1 4 32 10 SMAPE 128 50
∗ LR means the initial learning rate.

B E FFICIENCY A NALYSIS

In the main text, we have ploted the curve of efficiency in Figure 5. Here we present the quantitive
results in Table 8. It should be noted that TimeMixer’s outstanding efficiency advantage over
Transformer-based models, such as PatchTST, FEDformer, and Autoformer, is attributed to its fully
MLP-based network architecture.

Table 8: The GPU memory (MiB) and speed (running time, s/iter) of each model.
Series Length 192 384 768 1536 3072
Models Mem Speed Mem Speed Mem Speed Mem Speed Mem Speed
TimeMixer (Ours) 1003 0.007 1043 0.007 1075 0.008 1151 0.009 1411 0.016
PatchTST (2023) 1919 0.018 2097 0.019 2749 0.021 5465 0.032 16119 0.094
TimesNet (2023a) 1148 0.028 1245 0.024 1585 0.042 2491 0.045 2353 0.073
Crossformer (2023) 1737 0.027 1799 0.027 1895 0.028 2303 0.033 3759 0.035
MICN (2023) 1771 0.014 1801 0.016 1873 0.017 1991 0.018 2239 0.020
DLinear (2023) 1001 0.002 1021 0.003 1041 0.003 1081 0.0.004 1239 0.015
FEDFormer (2022b) 2567 0.132 5977 0.141 7111 0.143 9173 0.178 12485 0.288
Autoformer (2021) 1761 0.028 2101 0.070 3209 0.071 5395 0.129 10043 0.255

C E RROR BARS

In this paper, we repeat all the experiments three times. Here we report the standard deviation of our
model and the second best model, as well as the statistical significance test in Table 9, 10, 11.

D H YPERPARAMTER S ENSITIVITY

In the main text, we have explored the effect of number of scales M . Here, we further evaluate the
number of layers L. As shown in Table 12, we can find that in general, increasing the number of
layers (L) will bring improvements across different prediction lengths. Therefore, we set to 2 to trade
off efficiency and performance.

14
Published as a conference paper at ICLR 2024

Table 9: Standard deviation and statistical tests for our TimeMixer method and second-best method
(PatchTST) on ETT, Weather, Solar-Energy, Electricity and Traffic datasets.
Model TimeMixer PatchTST (2023) Confidence
Dataset MSE MAE MSE MAE Interval
Weather 0.240 ± 0.010 0.271 ± 0.009 0.265 ± 0.012 0.285 ± 0.011 99%
Solar-Energy 0.216 ± 0.002 0.280 ± 0.022 0.287 ± 0.020 0.333 ± 0.018 99%
Electricity 0.182 ± 0.017 0.272 ± 0.006 0.216 ± 0.012 0.318 ± 0.015 99%
Traffic 0.484 ± 0.015 0.297 ± 0.013 0.529 ± 0.008 0.341 ± 0.002 99%
ETTh1 0.447 ± 0.002 0.440 ± 0.005 0.516 ± 0.003 0.484 ± 0.005 99%
ETTh2 0.364 ± 0.008 0.395 ± 0.010 0.391 ± 0.005 0.411 ± 0.003 99%
ETTm1 0.381 ± 0.003 0.395 ± 0.006 0.400 ± 0.002 0.407 ± 0.005 99%
ETTm2 0.275 ± 0.001 0.323 ± 0.003 0.290 ± 0.002 0.334 ± 0.002 99%

Table 10: Standard deviation and statistical tests for our TimeMixer method and second-best method
(SCINet) on PEMS dataset.
Model TimeMixer SCINet (2022a) Confidence
Dataset MAE MAPE RMSE MAE MAPE RMSE Interval
PEMS03 14.63 ± 0.112 14.54 ± 0.105 23.28 ± 0.128 15.97 ± 0.153 15.89 ± 0.122 25.20 ± 0.137 99%
PEMS04 19.21 ± 0.217 12.53 ± 0.154 30.92 ± 0.143 20.35 ± 0.201 12.84 ± 0.213 32.31 ± 0.178 95%
PEMS07 20.57 ± 0.158 8.62 ± 0.112 33.59 ± 0.273 22.79 ± 0.179 9.41 ± 0.105 35.61 ± 0.112 99%
PEMS08 15.22 ± 0.311 9.67 ± 0.101 24.26 ± 0.212 17.38 ± 0.332 10.80 ± 0.219 27.34 ± 0.178 99%

E F ULL R ESULTS

To ensure a fair comparison between models, we conducted experiments using unified parameters
and reported results in the main text, including aligning all the input lengths, batch sizes, and training
epochs in all experiments. Here, we provide the full results for each forecasting setting in Table 13.
In addition, considering that the reported results in different papers are mostly obtained through
hyperparameter search, we provide the experiment results with the full version of the parameter
search. We searched for input length among 96, 192, 336, and 512, learning rate from 10−5 to 0.05,
encoder layers from 1 to 5, the dmodel from 16 to 512, training epochs from 10 to 100. The results are
included in Table 14, which can be used to compare the upper bound of each forecasting model.
We can find that the relative promotion of TimesMixer over PatchTST is smaller under comprehensive
hyperparameter search than the unified hyperparameter setting. It is worth noticing that TimeMixer
runs much faster than PatchTST according to the efficiency comparison in Table 8. Therefore,
considering perfromance, hyperparameter-search cost and efficiency, we believe TimeMixer is a
practical model in real-world applications and is valuable to deep time series forecasting community.

Table 11: Standard deviation and statistical tests for our TimeMixer method and second-best method
(TimesNet) on M4 dataset.
Model TimeMixer TimesNet (2023a) Confidence
Dataset SMAPE MAPE OWA SMAPE MAPE OWA Interval
Yearly 13.206 ± 0.121 2.916 ± 0.022 0.776 ± 0.002 13.387 ± 0.112 2.996 ± 0.017 0.786 ± 0.010 95%
Quarterly 9.996 ± 0.101 1.166 ± 0.015 0.825 ± 0.008 10.100 ± 0.105 1.182 ± 0.012 0.890 ± 0.006 95%
Monthly 12.605 ± 0.115 0.919 ± 0.011 0.869 ± 0.003 12.670 ± 0.106 0.933 ± 0.008 0.878 ± 0.001 95%
Others 4.564 ± 0.114 3.115 ± 0.027 0.982 ± 0.011 4.891 ± 0.120 3.302 ± 0.023 1.035 ± 0.017 99%
Averaged 11.723 ± 0.011 1.559 ± 0.022 0.840 ± 0.001 11.829 ± 0.120 1.585 ± 0.017 0.851 ± 0.003 95%

15
Published as a conference paper at ICLR 2024

Table 12: The MSE results of different number of scales (M ) and layers (L) on the ETTm1 dataset.
Predict Length Predict Length
96 192 336 720 96 192 336 720
Num. of Scales Num. of Layers
1 0.326 0.371 0.405 0.469 1 0.328 0.369 0.405 0.467
2 0.323 0.365 0.401 0.460 2 0.320 0.361 0.390 0.454
3 0.320 0.361 0.390 0.454 3 0.321 0.360 0.389 0.451
4 0.321 0.360 0.388 0.454 4 0.318 0.361 0.385 0.452
5 0.321 0.362 0.389 0.461 5 0.322 0.359 0.390 0.457

F F ULL A BLATIONS
Here we provide the complete results of ablations and alternative designs for TimeMixer.

F.1 A BLATIONS OF E ACH D ESIGN IN T IME M IXER

To verify the effectiveness of our design in TimeMixer, we conduct comprehensive ablations for all
benchmarks. All the results are provided in Table 15, 16, 17 as a supplement to Table 5 of main text.

Implementations We implement the following 10 types of ablations:

• Offical design in TimeMixer (case ①).

• Ablations on Future Mixing (case ②): In this case, we only adopt a single predictor to the
finest scale features, that is x̂ = Predictor0 (xL
0 ).
• Ablations on Past Mixing (case ③-⑦): Firstly, we remove the mixing operation of TimeMixer
in seasonal and trend parts respectively (case ③-④), that is removing Bottom-Up-Mixing
or Top-Down-Mixing layer. Then, we reverse the mixing directions for seasonal and
trend parts (case ⑤-⑦), which means adopting Bottom-Up-Mixing layer to trend and
Top-Down-Mixing layer to seasonal part.
• Ablations on Decomposition (case ⑧-⑩): In these cases, we do not adopt the decomposition,
which means that there is only one single feature for each scale. Thus, we can only try one
single mixing direction for these features, that is bottom-up mixing in case ⑧, top-down
mixing in case ⑨. Besides, we also test the case that is without mixing in ⑩, where the
interactions among multiscale features are removed.

Analysis In all ablations, we can find that the official design in TimeMixer performs best, which
provides solid support to our insights in special mixing approaches. Notably, it is observed that
completely reversing mixing directions for seasonal and trend parts (case ⑦) leads to a seriously per-
formance drop. This may come from that the essential microscopic information in finer-scale seasons
and macroscopic information in coarser-scale trends are ruined by unsuitable mixing approaches.

F.2 A LTERNATIVE D ECOMPOSITION M ETHODS

In this paper, we adopt the moving-average-based season-trend decomposition, which is widely

used in previous work, such as Autoformer (Wu et al., 2021), FEDformer (Zhou et al., 2022b) and
DLinear (Zeng et al., 2023). It is notable that Discrete Fourier Transformer (DFT) have been widely
recognized in time series analysis. Thus, we also try the DFT-based decomposition as a substitute.
Here we present two types of experiments.
The first one is DFT-based high- and low-frequency decomposition. We treat the high-frequency part
like the seasonal part in TimeMixer and the low-frequency part like the trend part. The results are
shown in Table 18. It observed that DFT-based decomposition performs worse than our design in
TimeMixer. Since we only explore the proper mixing approach for decomposed seasonal and trend
parts in the paper, the bottom-up and top-down mixing strategies may be not suitable for high- and
low-frequency parts. New visualizations like Figure 3 and 4 are expected to provide insights to the
model design. Thus, we would like to leave the exploration of DFT-based high- and low-frequency
decomposition methods as the future work.

16
Published as a conference paper at ICLR 2024

Table 13: Unified hyperparameter results for the long-term forecasting task. We compare extensive
competitive models under different prediction lengths. Avg is averaged from all four prediction
lengths, that is 96, 192, 336, 720.
TimeMixer PatchTST TimesNet Crossformer MICN FiLM DLinear FEDformer Stationary Autoformer Informer
Models
(Ours) 2023 2023a 2023 2023 2022a 2023 2022b 2022b 2021 2021

Metric MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE
96 0.163 0.209 0.186 0.227 0.172 0.220 0.195 0.271 0.198 0.261 0.195 0.236 0.195 0.252 0.217 0.296 0.173 0.223 0.266 0.336 0.300 0.384
Weather

192 0.208 0.250 0.234 0.265 0.219 0.261 0.209 0.277 0.239 0.299 0.239 0.271 0.237 0.295 0.276 0.336 0.245 0.285 0.307 0.367 0.598 0.544
336 0.251 0.287 0.284 0.301 0.246 0.337 0.273 0.332 0.285 0.336 0.289 0.306 0.282 0.331 0.339 0.380 0.321 0.338 0.359 0.395 0.578 0.523
720 0.339 0.341 0.356 0.349 0.365 0.359 0.379 0.401 0.351 0.388 0.361 0.351 0.345 0.382 0.403 0.428 0.414 0.410 0.419 0.428 1.059 0.741
Avg 0.240 0.271 0.265 0.285 0.251 0.294 0.264 0.320 0.268 0.321 0.271 0.291 0.265 0.315 0.309 0.360 0.288 0.314 0.338 0.382 0.634 0.548
Solar-Energy

96 0.189 0.259 0.265 0.323 0.373 0.358 0.232 0.302 0.257 0.325 0.333 0.350 0.290 0.378 0.286 0.341 0.321 0.380 0.456 0.446 0.287 0.323
192 0.222 0.283 0.288 0.332 0.397 0.376 0.371 0.410 0.278 0.354 0.371 0.372 0.320 0.398 0.291 0.337 0.346 0.369 0.588 0.561 0.297 0.341
336 0.231 0.292 0.301 0.339 0.420 0.380 0.495 0.515 0.298 0.375 0.408 0.385 0.353 0.415 0.354 0.416 0.357 0.387 0.595 0.588 0.367 0.429
720 0.223 0.285 0.295 0.336 0.420 0.381 0.526 0.542 0.299 0.379 0.406 0.377 0.357 0.413 0.380 0.437 0.375 0.424 0.733 0.633 0.374 0.431
Avg 0.216 0.280 0.287 0.333 0.403 0.374 0.406 0.442 0.283 0.358 0.380 0.371 0.330 0.401 0.328 0.383 0.350 0.390 0.586 0.557 0.331 0.381
96 0.153 0.247 0.190 0.296 0.168 0.272 0.219 0.314 0.180 0.293 0.198 0.274 0.210 0.302 0.193 0.308 0.169 0.273 0.201 0.317 0.274 0.368
Electricity

192 0.166 0.256 0.199 0.304 0.184 0.322 0.231 0.322 0.189 0.302 0.198 0.278 0.210 0.305 0.201 0.315 0.182 0.286 0.222 0.334 0.296 0.386
336 0.185 0.277 0.217 0.319 0.198 0.300 0.246 0.337 0.198 0.312 0.217 0.300 0.223 0.319 0.214 0.329 0.200 0.304 0.231 0.443 0.300 0.394
720 0.225 0.310 0.258 0.352 0.220 0.320 0.280 0.363 0.217 0.330 0.278 0.356 0.258 0.350 0.246 0.355 0.222 0.321 0.254 0.361 0.373 0.439
Avg 0.182 0.272 0.216 0.318 0.193 0.304 0.244 0.334 0.196 0.309 0.223 0.302 0.225 0.319 0.214 0.327 0.193 0.296 0.227 0.338 0.311 0.397
96 0.462 0.285 0.526 0.347 0.593 0.321 0.644 0.429 0.577 0.350 0.647 0.384 0.650 0.396 0.587 0.366 0.612 0.338 0.613 0.388 0.719 0.391
192 0.473 0.296 0.522 0.332 0.617 0.336 0.665 0.431 0.589 0.356 0.600 0.361 0.598 0.370 0.604 0.373 0.613 0.340 0.616 0.382 0.696 0.379
Traffic

336 0.498 0.296 0.517 0.334 0.629 0.336 0.674 0.420 0.594 0.358 0.610 0.367 0.605 0.373 0.621 0.383 0.618 0.328 0.622 0.337 0.777 0.420
720 0.506 0.313 0.552 0.352 0.640 0.350 0.683 0.424 0.613 0.361 0.691 0.425 0.645 0.394 0.626 0.382 0.653 0.355 0.660 0.408 0.864 0.472
Avg 0.484 0.297 0.529 0.341 0.620 0.336 0.667 0.426 0.593 0.356 0.637 0.384 0.625 0.383 0.610 0.376 0.624 0.340 0.628 0.379 0.764 0.416
96 0.375 0.400 0.460 0.447 0.384 0.402 0.423 0.448 0.426 0.446 0.438 0.433 0.397 0.412 0.395 0.424 0.513 0.491 0.449 0.459 0.865 0.713
ETTh1

192 0.429 0.421 0.512 0.477 0.436 0.429 0.471 0.474 0.454 0.464 0.493 0.466 0.446 0.441 0.469 0.470 0.534 0.504 0.500 0.482 1.008 0.792
336 0.484 0.458 0.546 0.496 0.638 0.469 0.570 0.546 0.493 0.487 0.547 0.495 0.489 0.467 0.530 0.499 0.588 0.535 0.521 0.496 1.107 0.809
720 0.498 0.482 0.544 0.517 0.521 0.500 0.653 0.621 0.526 0.526 0.586 0.538 0.513 0.510 0.598 0.544 0.643 0.616 0.514 0.512 1.181 0.865
Avg 0.447 0.440 0.516 0.484 0.495 0.450 0.529 0.522 0.475 0.480 0.516 0.483 0.461 0.457 0.498 0.484 0.570 0.537 0.496 0.487 1.040 0.795
96 0.289 0.341 0.308 0.355 0.340 0.374 0.745 0.584 0.372 0.424 0.322 0.364 0.340 0.394 0.358 0.397 0.476 0.458 0.346 0.388 3.755 1.525
ETTh2

192 0.372 0.392 0.393 0.405 0.402 0.414 0.877 0.656 0.492 0.492 0.404 0.414 0.482 0.479 0.429 0.439 0.512 0.493 0.456 0.452 5.602 1.931
336 0.386 0.414 0.427 0.436 0.452 0.452 1.043 0.731 0.607 0.555 0.435 0.445 0.591 0.541 0.496 0.487 0.552 0.551 0.482 0.486 4.721 1.835
720 0.412 0.434 0.436 0.450 0.462 0.468 1.104 0.763 0.824 0.655 0.447 0.458 0.839 0.661 0.463 0.474 0.562 0.560 0.515 0.511 3.647 1.625
Avg 0.364 0.395 0.391 0.411 0.414 0.427 0.942 0.684 0.574 0.531 0.402 0.420 0.563 0.519 0.437 0.449 0.526 0.516 0.450 0.459 4.431 1.729
96 0.320 0.357 0.352 0.374 0.338 0.375 0.404 0.426 0.365 0.387 0.353 0.370 0.346 0.374 0.379 0.419 0.386 0.398 0.505 0.475 0.672 0.571
ETTm1

192 0.361 0.381 0.390 0.393 0.374 0.387 0.450 0.451 0.403 0.408 0.389 0.387 0.382 0.391 0.426 0.441 0.459 0.444 0.553 0.496 0.795 0.669
336 0.390 0.404 0.421 0.414 0.410 0.411 0.532 0.515 0.436 0.431 0.421 0.408 0.415 0.415 0.445 0.459 0.495 0.464 0.621 0.537 1.212 0.871
720 0.454 0.441 0.462 0.449 0.478 0.450 0.666 0.589 0.489 0.462 0.481 0.441 0.473 0.451 0.543 0.490 0.585 0.516 0.671 0.561 1.166 0.823
Avg 0.381 0.395 0.406 0.407 0.400 0.406 0.513 0.495 0.423 0.422 0.411 0.402 0.404 0.408 0.448 0.452 0.481 0.456 0.588 0.517 0.961 0.734
96 0.175 0.258 0.183 0.270 0.187 0.267 0.287 0.366 0.197 0.296 0.183 0.266 0.193 0.293 0.203 0.287 0.192 0.274 0.255 0.339 0.365 0.453
ETTm2

192 0.237 0.299 0.255 0.314 0.249 0.309 0.414 0.492 0.284 0.361 0.248 0.305 0.284 0.361 0.269 0.328 0.280 0.339 0.281 0.340 0.533 0.563
336 0.298 0.340 0.309 0.347 0.321 0.351 0.597 0.542 0.381 0.429 0.309 0.343 0.382 0.429 0.325 0.366 0.334 0.361 0.339 0.372 1.363 0.887
720 0.391 0.396 0.412 0.404 0.408 0.403 1.730 1.042 0.549 0.522 0.410 0.400 0.558 0.525 0.421 0.415 0.417 0.413 0.433 0.432 3.379 1.338
Avg 0.275 0.323 0.290 0.334 0.291 0.333 0.757 0.610 0.353 0.402 0.287 0.329 0.354 0.402 0.305 0.349 0.306 0.347 0.327 0.371 1.410 0.810

17
Published as a conference paper at ICLR 2024

Table 14: Experiment results under hyperparameter searching for the long-term forecasting task. Avg
is averaged from all four prediction lengths.
TimeMixer PatchTST TimesNet Crossformer MICN FiLM DLinear FEDformer Stationary Autoformer Informer
Models
(Ours) 2023 2023a 2023 2023 2022a 2023 2022b 2022b 2021 2021

Metric MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE
96 0.147 0.197 0.149 0.198 0.172 0.220 0.232 0.302 0.161 0.229 0.199 0.262 0.176 0.237 0.217 0.296 0.173 0.223 0.266 0.336 0.300 0.384
Weather

192 0.189 0.239 0.194 0.241 0.219 0.261 0.371 0.410 0.220 0.281 0.228 0.288 0.220 0.282 0.276 0.336 0.245 0.285 0.307 0.367 0.598 0.544
336 0.241 0.280 0.306 0.282 0.246 0.337 0.495 0.515 0.278 0.331 0.267 0.323 0.265 0.319 0.339 0.380 0.321 0.338 0.359 0.395 0.578 0.523
720 0.310 0.330 0.314 0.334 0.365 0.359 0.526 0.542 0.311 0.356 0.319 0.361 0.323 0.362 0.403 0.428 0.414 0.410 0.419 0.428 1.059 0.741

Avg 0.222 0.262 0.241 0.264 0.251 0.294 0.406 0.442 0.242 0.299 0.253 0.309 0.246 0.300 0.309 0.360 0.288 0.314 0.338 0.382 0.634 0.548
Solar-Energy

96 0.167 0.220 0.224 0.278 0.219 0.314 0.181 0.240 0.188 0.252 0.320 0.339 0.289 0.377 0.201 0.304 0.321 0.380 0.456 0.446 0.200 0.247
192 0.187 0.249 0.253 0.298 0.231 0.322 0.196 0.252 0.215 0.280 0.360 0.362 0.319 0.397 0.237 0.337 0.346 0.369 0.588 0.561 0.220 0.251
336 0.200 0.258 0.273 0.306 0.246 0.337 0.216 0.243 0.222 0.267 0.398 0.375 0.352 0.415 0.254 0.362 0.357 0.387 0.595 0.588 0.260 0.287
720 0.215 0.250 0.272 0.308 0.280 0.363 0.220 0.256 0.226 0.264 0.399 0.368 0.356 0.412 0.280 0.397 0.335 0.384 0.733 0.633 0.244 0.301

Avg 0.192 0.244 0.256 0.298 0.244 0.334 0.204 0.248 0.213 0.266 0.369 0.361 0.329 0.400 0.243 0.350 0.340 0.380 0.593 0.557 0.231 0.272

96 0.129 0.224 0.129 0.222 0.168 0.272 0.150 0.251 0.164 0.269 0.154 0.267 0.140 0.237 0.193 0.308 0.169 0.273 0.201 0.317 0.274 0.368
Electricity

192 0.140 0.220 0.147 0.240 0.184 0.322 0.161 0.260 0.177 0.285 0.164 0.258 0.153 0.249 0.201 0.315 0.182 0.286 0.222 0.334 0.296 0.386
336 0.161 0.255 0.163 0.259 0.198 0.300 0.182 0.281 0.193 0.304 0.188 0.283 0.169 0.267 0.214 0.329 0.200 0.304 0.231 0.338 0.300 0.394
720 0.194 0.287 0.197 0.290 0.220 0.320 0.251 0.339 0.212 0.321 0.236 0.332 0.203 0.301 0.246 0.355 0.222 0.321 0.254 0.361 0.373 0.439

Avg 0.156 0.246 0.159 0.253 0.192 0.295 0.186 0.283 0.186 0.295 0.186 0.285 0.166 0.264 0.214 0.321 0.213 0.296 0.227 0.338 0.311 0.397

96 0.360 0.249 0.360 0.249 0.593 0.321 0.514 0.267 0.519 0.309 0.416 0.294 0.410 0.282 0.587 0.366 0.612 0.338 0.613 0.388 0.719 0.391
Traffic

192 0.375 0.250 0.379 0.256 0.617 0.336 0.549 0.252 0.537 0.315 0.408 0.288 0.423 0.287 0.604 0.373 0.613 0.340 0.616 0.382 0.696 0.379
336 0.385 0.270 0.392 0.264 0.629 0.336 0.530 0.300 0.534 0.313 0.425 0.298 0.436 0.296 0.621 0.383 0.618 0.328 0.622 0.337 0.777 0.420
720 0.430 0.281 0.432 0.286 0.640 0.350 0.573 0.313 0.577 0.325 0.520 0.353 0.466 0.315 0.626 0.382 0.653 0.355 0.660 0.408 0.864 0.472

Avg 0.387 0.262 0.391 0.264 0.620 0.336 0.542 0.283 0.541 0.315 0.442 0.308 0.434 0.295 0.609 0.376 0.624 0.340 0.628 0.379 0.764 0.415

96 0.361 0.390 0.370 0.400 0.384 0.402 0.418 0.438 0.421 0.431 0.422 0.432 0.375 0.399 0.376 0.419 0.513 0.491 0.449 0.459 0.865 0.713
ETTh1

192 0.409 0.414 0.413 0.429 0.436 0.429 0.539 0.517 0.474 0.487 0.462 0.458 0.405 0.416 0.420 0.448 0.534 0.504 0.500 0.482 1.008 0.792
336 0.430 0.429 0.422 0.440 0.638 0.469 0.709 0.638 0.569 0.551 0.501 0.483 0.439 0.443 0.459 0.465 0.588 0.535 0.521 0.496 1.107 0.809
720 0.445 0.460 0.447 0.468 0.521 0.500 0.733 0.636 0.770 0.672 0.544 0.526 0.472 0.490 0.506 0.507 0.643 0.616 0.514 0.512 1.181 0.865

Avg 0.411 0.423 0.413 0.434 0.458 0.450 0.600 0.557 0.558 0.535 0.482 0.475 0.423 0.437 0.440 0.460 0.57 0.536 0.496 0.487 1.040 0.795

96 0.271 0.330 0.274 0.337 0.340 0.374 0.425 0.463 0.299 0.364 0.323 0.370 0.289 0.353 0.346 0.388 0.476 0.458 0.358 0.397 3.755 1.525
ETTh2

192 0.317 0.402 0.314 0.382 0.231 0.322 0.473 0.500 0.441 0.454 0.391 0.415 0.383 0.418 0.429 0.439 0.512 0.493 0.456 0.452 5.602 1.931
336 0.332 0.396 0.329 0.384 0.452 0.452 0.581 0.562 0.654 0.567 0.415 0.440 0.448 0.465 0.496 0.487 0.552 0.551 0.482 0.486 4.721 1.835
720 0.342 0.408 0.379 0.422 0.462 0.468 0.775 0.665 0.956 0.716 0.441 0.459 0.605 0.551 0.463 0.474 0.562 0.560 0.515 0.511 3.647 1.625

Avg 0.316 0.384 0.324 0.381 0.371 0.404 0.564 0.548 0.588 0.525 0.393 0.421 0.431 0.447 0.433 0.447 0.526 0.516 0.453 0.462 4.431 1.729

96 0.291 0.340 0.293 0.346 0.338 0.375 0.361 0.403 0.316 0.362 0.302 0.345 0.299 0.343 0.379 0.419 0.386 0.398 0.505 0.475 0.672 0.571
ETTm1

192 0.327 0.365 0.333 0.370 0.374 0.387 0.387 0.422 0.363 0.390 0.338 0.368 0.335 0.365 0.426 0.441 0.459 0.444 0.553 0.496 0.795 0.669
336 0.360 0.381 0.369 0.392 0.410 0.411 0.605 0.572 0.408 0.426 0.373 0.388 0.369 0.386 0.445 0.459 0.495 0.464 0.621 0.537 1.212 0.871
720 0.415 0.417 0.416 0.420 0.478 0.450 0.703 0.645 0.481 0.476 0.420 0.420 0.425 0.421 0.543 0.490 0.585 0.516 0.671 0.561 1.166 0.823

Avg 0.348 0.375 0.353 0.382 0.353 0.382 0.514 0.510 0.392 0.413 0.358 0.38 0.357 0.379 0.448 0.452 0.481 0.456 0.588 0.517 0.961 0.733

96 0.164 0.254 0.166 0.256 0.187 0.267 0.275 0.358 0.179 0.275 0.165 0.256 0.167 0.260 0.203 0.287 0.192 0.274 0.255 0.339 0.365 0.453
ETTm2

192 0.223 0.295 0.223 0.296 0.249 0.309 0.345 0.400 0.307 0.376 0.222 0.296 0.224 0.303 0.269 0.328 0.280 0.339 0.281 0.340 0.533 0.563
336 0.279 0.330 0.274 0.329 0.321 0.351 0.657 0.528 0.325 0.388 0.277 0.333 0.281 0.342 0.325 0.366 0.334 0.361 0.339 0.372 1.363 0.887
720 0.359 0.383 0.362 0.385 0.408 0.403 1.208 0.753 0.502 0.490 0.371 0.389 0.397 0.421 0.421 0.415 0.417 0.413 0.422 0.419 3.379 1.388

Avg 0.256 0.315 0.256 0.317 0.291 0.333 0.621 0.510 0.328 0.382 0.259 0.319 0.267 0.332 0.304 0.349 0.306 0.347 0.324 0.368 1.410 0.823

18
Published as a conference paper at ICLR 2024

Table 15: Ablations on both Past-Decompose-Mixing (Decompose, Season Mixing, and Trend Mixing)
and Future-Multipredictor-Mixing blocks in predict-336 setting for all long-term benchmarks. ↗
indicates the bottom-up mixing while ↙ indicates top-down. A check mark ✓ and a wrong mark ×
indicate with and without certain components respectively. ① is the official design in TimeMixer.
Past mixing Future mixing ETTh1 ETTh2 ETTm1 ETTm2 Weather Solar Electricity Traffic
Case Decompose
Seasonal Trend Multipredictor MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE
① ✓ ↗ ↙ ✓ 0.484 0.458 0.386 0.414 0.390 0.404 0.298 0.340 0.251 0.287 0.231 0.292 0.185 0.277 0.498 0.296
② ✓ ↗ ↙ × 0.493 0.472 0.399 0.426 0.402 0.415 0.311 0.357 0.262 0.308 0.267 0.339 0.198 0.301 0.518 0.337
③ ✓ × ↙ ✓ 0.507 0.490 0.408 0.437 0.411 0.427 0.322 0.366 0.273 0.321 0.274 0.355 0.207 0.304 0.532 0.348
④ ✓ ↗ × ✓ 0.491 0.483 0.397 0.424 0.405 0.414 0.317 0.351 0.269 0.311 0.268 0.341 0.200 0.299 0.525 0.339
⑤ ✓ ↙ ↙ ✓ 0.488 0.466 0.393 0.426 0.392 0.413 0.309 0.349 0.257 0.293 0.252 0.330 0.191 0.293 0.520 0.331
⑥ ✓ ↗ ↗ ✓ 0.493 0.484 0.401 0.432 0.396 0.415 0.319 0.361 0.271 0.322 0.281 0.363 0.214 0.307 0.541 0.351
⑦ ✓ ↙ ↗ ✓ 0.498 0.491 0.421 0.436 0.412 0.429 0.321 0.369 0.277 0.332 0.298 0.375 0.221 0.319 0.564 0.357
⑧ × ↗ ✓ 0.494 0.488 0.396 0.421 0.395 0.408 0.313 0.360 0.259 0.308 0.260 0.321 0.199 0.303 0.522 0.340
⑨ × ↙ ✓ 0.487 0.462 0.394 0.419 0.393 0.406 0.307 0.354 0.261 0.327 0.257 0.334 0.196 0.310 0.526 0.339
⑩ × × ✓ 0.502 0.489 0.411 0. 427 0.405 0.412 0.319 0.358 0.273 0.331 0.295 0.336 0.217 0.318 0.558 0.347

Table 16: Ablations on both Past-Decompose-Mixing (Decompose, Season Mixing, and Trend Mixing)
and Future-Multipredictor-Mixing blocks in the M4 short-term forecasting benchmark. Case notations
are same as Table 5 and 15. ① is the official design in TimeMixer.
Past mixing Future mixing M4
Case Decompose
Seasonal Trend Multipredictor SMAPE MASE OWA
① ✓ ↗ ↙ ✓ 11.723 1.559 0.840
② ✓ ↗ ↙ × 12.503 1.634 0.925
③ ✓ × ↙ ✓ 13.051 1.676 0.962
④ ✓ ↗ × ✓ 12.911 1.655 0.941
⑤ ✓ ↙ ↙ ✓ 12.008 1.628 0.871
⑥ ✓ ↗ ↗ ✓ 11.978 1.626 0.859
⑦ ✓ ↙ ↗ ✓ 13.012 1.657 0.954
⑧ × ↗ ✓ 11.975 1.617 0.851
⑨ × ↙ ✓ 11.973 1.622 0.850
⑩ × × ✓ 12.468 1.671 0.916

The second one is to enhance season-trend decomposition with DFT. Here we present the DFT-based
season-trend decomposition. Firstly, we transform the raw series into a frequency domain by DFT
and then extract the most significant frequencies. After transforming the selected frequencies by
inverse DFT, we obtain the seasonal part of the time series. Then the trend part is the raw series minus
the seasonal part. We can find that this superior decomposition method surpasses the moving-average
design. However, since moving average is quite simple and easy to implement with PyTorch, we
eventually chose the moving-average-based season-trend decomposition in TimeMixer, which can
also achieve a favorable balance between performance and efficiency.

Table 17: Ablations on both Past-Decompose-Mixing (Decompose, Season Mixing, and Trend Mixing)
and Future-Multipredictor-Mixing blocks in the PEMS short-term forecasting benchmarks. Case
notations are same as Table 5 and 15. ① is the official design in TimeMixer.
Past mixing Future mixing PEMS03 PEMS04 PEMS07 PEMS08
Case Decompose
Seasonal Trend Multipredictor MAE MAPE RMSE MAE MAPE RMSE MAE MAPE RMSE MAE MAPE RMSE
① ✓ ↗ ↙ ✓ 14.63 14.54 23.28 19.21 12.53 30.92 20.57 8.62 33.59 15.22 9.67 24.26
② ✓ ↗ ↙ × 15.66 15.81 25.77 21.67 13.45 34.89 22.78 9.52 35.57 17.48 10.91 27.84
③ ✓ × ↙ ✓ 18.90 17.33 30.75 24.49 16.28 38.79 25.27 10.74 40.06 19.02 11.71 30.05
④ ✓ ↗ × ✓ 17.67 17.58 28.48 22.91 15.02 37.14 24.81 10.02 38.68 18.29 12.21 28.62
⑤ ✓ ↙ ↙ ✓ 15.46 15.73 24.91 20.78 13.02 32.47 22.57 9.33 35.87 16.54 9.97 26.88
⑥ ✓ ↗ ↗ ✓ 15.32 15.41 24.83 21.09 13.78 33.11 21.94 9.41 35.40 17.01 10.82 26.93
⑦ ✓ ↙ ↗ ✓ 18.81 17.29 29.78 22.27 15.14 34.67 25.11 10.60 39.74 18.74 12.09 28.67
⑧ × ↗ ✓ 15.57 15.62 24.98 21.51 13.47 34.81 22.94 9.81 35.49 18.17 11.02 28.14
⑨ × ↙ ✓ 15.48 15.55 24.83 21.79 14.03 35.23 21.93 9.91 36.02 17.71 10.88 27.91
⑩ × × ✓ 19.01 18.58 30.06 24.87 16.66 39.48 24.72 9.97 37.18 19.18 12.21 30.79

19
Published as a conference paper at ICLR 2024

Table 18: Alternative decomposition methods in M4, PEMS04 and predict-336 setting of ETTm1.
ETTm1 M4 PEMS04
Decomposition methods
MSE MAE SMAPE MASE OWA MAE MAPE RMSE
DFT-based high- and low-frequency decomposition 0.392 0.404 12.054 1.632 0.862 19.83 12.74 31.48
DFT-based season-trend decomposition 0.383 0.399 11.673 1.536 0.824 18.91 12.27 29.47
Moving-average-based season-trend decomposition (TimeMixer) 0.390 0.404 11.723 1.559 0.840 19.21 12.53 30.92

F.3 A LTERNATIVE D OWNSAMPLING M ETHODS

As we stated in Section 3.1, we adopt the average pooling to obtain the multiscale series. Here
we replace this operation with 1D convolutions. From Table 19, we can find that the complicated
1D-convolution-based outperforms average pooling slightly. But considering both performance and
efficiency, we eventually use average pooling in TimeMixer.

Table 19: Alternative downsampling methods in M4, PEMS04 and predict-336 setting of ETTm1.
ETTm1 M4 PEMS04
Downsampling methods
MSE MAE SMAPE MASE OWA MAE MAPE RMSE
Moving average 0.390 0.404 11.723 1.559 0.840 19.21 12.53 30.92
1D convolutions with stride as 2 0.387 0.401 11.682 1.542 0.831 19.04 12.17 29.88

F.4 A LTERNATIVE E NSEMBLE S TRATEGIES

In the main text, we sum the outputs from multiple predictors towards the final result (Eq. 6). Here
we also try the average strategy. Note that in TimeMixer, the loss is calculated based on the ensemble
PM
results, not for each predictor, that is ∥x − x̂∥ = ∥x − m=0 x̂m ∥. When we change the ensemble
PM
strategy as average, the loss will be ∥x − x̂∥ = ∥x − M1+1 m=0 x̂m ∥. Obviously, the difference
between average and mean strategies is only a constant multiple.
It is common sense in the deep learning community that deep models can easily fit constant multiple.
For example, if we replace the “sum” with “average”, under the same supervision, the deep model
can easily fit this change by learning the parameters of each predictor equal to the M1+1 of the “sum”
case, which means these two designs are equivalent in learning the final prediction under the deep
model aspect. Besides, we also provide the experiment results in Table 20, where we can find that the
performances of these two strategies are almost the same.

Table 20: Alternative ensemble strategies in M4, PEMS04 and predict-336 setting of ETTm1.
ETTm1 M4 PEMS04
Ensemble strategies
MSE MAE SMAPE MASE OWA MAE MAPE RMSE
Sum ensemble 0.390 0.404 11.723 1.559 0.840 19.21 12.53 30.92
Average ensemble 0.391 0.407 11.742 1.573 0.851 19.17 12.45 30.88

F.5 A BLATIONS ON L ARGER S CALES AND L ARGER I NPUT L ENGTH S ETTINGS

In the previous section (Table 15, 16, 17), we have conducted comprehensive ablations under the
unified configuration presented in Table 7, which M is set to 1 for short-term forecasting and input
length is set to 96 for short-term forecasting. To further evaluate the effectiveness of our proposed
module, we also provide additional ablations on larger scales for short-term forecasting and larger
input length settings in Table 21 as a supplement to Table 5 of the main text. Besides, we also provide
a detailed analysis of relative promotion (Table 22), where we can find the following observations:

20
Published as a conference paper at ICLR 2024

Table 21: A supplement to Table 5 of the main text with ablations on both PDM (Decompose, Season
Mixing, Trend Mixing) and FMM blocks in PEMS04 with M = 3 and predict-336 setting of ETTm1
with input-336. Since the input length of M4 is fixed to a small value, larger M may result in a
meaningless configuration. We only experiment on PEMS04 here.
Past mixing Future mixing PEMS04 with M = 3 ETTm1 with input 336
Case Decompose
Seasonal Trend Multipredictor MAE MAPE RMSE MSE MAE
① ✓ ↗ ↙ ✓ 18.10 11.73 28.51 0.360 0.381
② ✓ ↗ ↙ × 21.49 13.12 33.48 0.375 0.398
③ ✓ × ↙ ✓ 23.68 16.01 37.42 0.390 0.415
④ ✓ ↗ × ✓ 22.44 14.81 36.54 0.386 0.410
⑤ ✓ ↙ ↙ ✓ 20.41 13.08 31.92 0.371 0.389
⑥ ✓ ↗ ↗ ✓ 21.28 13.19 32.84 0.370 0.388
⑦ ✓ ↙ ↗ ✓ 22.16 14.60 35.42 0.384 0.409
⑧ × ↗ ✓ 20.98 13.12 33.94 0.372 0.396
⑨ × ↙ ✓ 20.66 13.06 32.74 0.374 0.398
⑩ × × ✓ 24.16 16.21 38.04 0.401 0.414

Table 22: Relative promotion analysis on ablations under different hyperparameter configurations.
For example, the relative promotion is calculated by (1−①/②) in case ②.
Past mixing Future mixing PEMS04 with M = 3 ETTm1 with input 336
Case Decompose
Seasonal Trend Multipredictor MAE MAPE RMSE MSE MAE
① ✓ ↗ ↙ ✓ - - - - -
② ✓ ↗ ↙ × 15.8% 10.6% 14.9% 4.1% 4.2%
③ ✓ × ↙ ✓ 23.6% 26.7% 23.8% 7.7% 8.2%
④ ✓ ↗ × ✓ 19.2% 20.1% 22.0% 6.8% 7.1%
⑤ ✓ ↙ ↙ ✓ 11.4% 10.4% 10.7% 3.0% 2.1%
⑥ ✓ ↗ ↗ ✓ 15.0% 11.1% 13.2% 2.7% 1.8%
⑦ ✓ ↙ ↗ ✓ 18.4% 19.7% 19.5% 6.2% 6.9%
⑧ × ↗ ✓ 13.7% 10.6% 15.9% 3.2% 3.8%
⑨ × ↙ ✓ 12.4% 10.2% 13.0% 3.7% 4.2%
⑩ × × ✓ 25.1% 27.6% 25.0% 10.2% 8.0%

Past mixing Future mixing PEMS04 with M = 1 ETTm1 with input 96

Case Decompose
Seasonal Trend Multipredictor MAE MAPE RMSE MSE MAE
① ✓ ↗ ↙ ✓ - - - - -
② ✓ ↗ ↙ × 11.4% 6.8% 11.2% 2.9% 2.6%
③ ✓ × ↙ ✓ 21.6% 23.1% 20.2% 5.1% 5.4%
④ ✓ ↗ × ✓ 16.2% 16.6% 16.7% 3.7% 2.4%
⑤ ✓ ↙ ↙ ✓ 7.6% 3.7% 4.7% 0.5% 2.1%
⑥ ✓ ↗ ↗ ✓ 8.9% 9.0% 6.6% 1.5% 2.6%
⑦ ✓ ↙ ↗ ✓ 13.7% 17.2% 10.8% 5.3% 5.8%
⑧ × ↗ ✓ 10.6% 6.9% 11.7% 1.2% 1.0%
⑨ × ↙ ✓ 11.8% 10.6% 12.2% 0.7% 0.5%
⑩ × × ✓ 22.7% 24.7% 21.6% 3.9% 2.1%

• All the designs of TimeMixer are effective in both hyperparameter settings. Especially, the
seasonal mixing (case ③) and proper mixing directions (case ⑦) are essential.

21
Published as a conference paper at ICLR 2024

• As shown in Table 22, in large M and longer input length situations, the relative promotions
brought by seasonal and trend mixing in PDM and FMM are more significant in most cases,
which further verifies the effectiveness of our design.

• It is observed that the seasonal mixing direction contribution (case ⑤) is much more
significant in the longer-input setting on the ETTm1 dataset. This may come from that input-
96 only corresponds to one day in 15-minutely sampled ETTm1, while input-336 maintains
3.5 days of information (around 3.5 periods). Thus, the bottom-up mixing direction will
benefit from sufficient microscopic seasonal information under the longer-input setting.

G A DDITIONAL BASELINES

Due to the limitation of the main text, we also include three advanced baselines here: the general multi-
scale framework Scaleformer (Shabani et al., 2023), two concurrent MLP-based model MTSMixer
(Li et al., 2023) and TSMixer (Chen et al., 2023). Since the latter two baselines were not officially
published during our submission, we adopted their public code and reproduced them with both unified
hyperparameter setting and the hyperparameter searching settings. As presented in Table 23, 24, 25,
TimeMixer still performs best in comparison with these baselines. Showcases of these additional
baselines are also provided in Appendix I for an intuitive comparison.

H S PECTRAL A NALYSIS OF M ODEL P REDICTIONS

To demonstrate the advancement of TimeMixer, we plot the spectrum of ground truth and model
predictions. It is observed that TimeMixer captures different frequency parts precisely.

(a) Ground Truth spectrogram (b) TimeMixer spectrogram (c) PatchTST spectrogram

Figure 8: Prediction spectrogram cases from ETTh1 by ground truth and different models under the
input-96-predict-96 settings.

I S HOWCASES

In order to evaluate the performance of different models, we conduct the qualitative compari-
son by plotting the final dimension of forecasting results from the test set of each dataset (Fig-
ures 9, 10, 11, 12, 13, 17, 18). Among the various models, TimeMixer exhibits superior performance.

22
Published as a conference paper at ICLR 2024

Table 23: Unified and searched hyperparameter results for additional baselines in the long-term
forecasting task. We compare extensive competitive models under different prediction lengths. Avg is
averaged from all four prediction lengths, that is 96, 192, 336, 720. All these baselines are reproduced
by their official code. For the searched hyperparameter setting, we follow the searching strategy
described in Appendix E. Especially for TSMixer, we reproduced it in Pytorch (Paszke et al., 2019).
Unified Hyperparameter Searched Hyperparameter
Models
TimeMixer Scaleformer MTSMixer TSMixer TimeMixer Scaleformer MTSMixer TSMixer
(Ours) (2023) (2023) (2023) (Ours) (2023) (2023) (2023)

Metric MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE
96 0.163 0.209 0.220 0.289 0.173 0.224 0.175 0.247 0.147 0.197 0.192 0.241 0.167 0.221 0.149 0.198
Weather

192 0.208 0.250 0.341 0.385 0.219 0.261 0.224 0.294 0.189 0.239 0.220 0.288 0.208 0.250 0.201 0.251
336 0.251 0.287 0.463 0.455 0.274 0.300 0.262 0.326 0.241 0.280 0.288 0.324 0.298 0.302 0.287 0.291
720 0.339 0.341 0.640 0.565 0.365 0.359 0.349 0.348 0.310 0.330 0.321 0.360 0.344 0.339 0.320 0.336
Avg 0.240 0.271 0.416 0.423 0.258 0.286 0.253 0.304 0.222 0.262 0.248 0.304 0.254 0.278 0.240 0.269
Solar-Energy

96 0.189 0.259 0.271 0.331 0.217 0.272 0.216 0.294 0.167 0.220 0.224 0.328 0.199 0.251 0.190 0.272
192 0.222 0.283 0.288 0.332 0.258 0.299 0.294 0.359 0.187 0.249 0.247 0.312 0.221 0.275 0.207 0.278
336 0.231 0.292 0.358 0.412 0.278 0.310 0.302 0.367 0.200 0.258 0.274 0.308 0.231 0.281 0.246 0.304
720 0.223 0.285 0.377 0.437 0.293 0.321 0.311 0.372 0.215 0.250 0.335 0.384 0.270 0.301 0.274 0.308
Avg 0.216 0.280 0.323 0.378 0.261 0.300 0.280 0.348 0.192 0.244 0.270 0.333 0.230 0.277 0.229 0.283
96 0.153 0.247 0.182 0.297 0.173 0.270 0.190 0.299 0.129 0.224 0.162 0.274 0.154 0.267 0.142 0.234
Electricity

192 0.166 0.256 0.188 0.300 0.186 0.280 0.216 0.323 0.140 0.220 0.171 0.284 0.168 0.272 0.154 0.248
336 0.185 0.277 0.210 0.324 0.204 0.297 0.226 0.334 0.161 0.255 0.192 0.304 0.182 0.281 0.161 0.262
720 0.225 0.310 0.232 0.339 0.241 0.326 0.250 0.353 0.194 0.287 0.238 0.332 0.212 0.321 0.209 0.304
Avg 0.182 0.272 0.203 0.315 0.201 0.293 0.220 0.327 0.156 0.246 0.191 0.298 0.179 0.286 0.167 0.262
96 0.462 0.285 0.564 0.351 0.523 0.357 0.499 0.344 0.360 0.249 0.409 0.281 0.514 0.338 0.398 0.272
192 0.473 0.296 0.570 0.349 0.535 0.367 0.540 0.370 0.375 0.250 0.418 0.294 0.519 0.351 0.402 0.281
Traffic

336 0.498 0.296 0.576 0.349 0.566 0.379 0.557 0.378 0.385 0.270 0.427 0.294 0.557 0.361 0.412 0.294
720 0.506 0.313 0.602 0.360 0.608 0.397 0.586 0.397 0.430 0.281 0.518 0.356 0.569 0.362 0.448 0.311
Avg 0.484 0.297 0.578 0.352 0.558 0.375 0.546 0.372 0.387 0.262 0.443 0.307 0.539 0.354 0.415 0.290
96 0.375 0.400 0.401 0.428 0.418 0.437 0.387 0.411 0.361 0.390 0.381 0.412 0.397 0.428 0.370 0.402
ETTh1

192 0.429 0.421 0.471 0.478 0.463 0.460 0.441 0.437 0.409 0.414 0.445 0.441 0.452 0.466 0.406 0.414
336 0.484 0.458 0.527 0.498 0.516 0.478 0.507 0.467 0.430 0.429 0.501 0.484 0.487 0.462 0.424 0.434
720 0.498 0.482 0.578 0.547 0.532 0.549 0.527 0.548 0.445 0.460 0.544 0.528 0.510 0.506 0.471 0.479
Avg 0.447 0.440 0.495 0.488 0.482 0.481 0.466 0.467 0.411 0.423 0.468 0.466 0.461 0.464 0.418 0.432
96 0.289 0.341 0.368 0.398 0.343 0.378 0.308 0.357 0.271 0.330 0.340 0.394 0.328 0.367 0.271 0.339
ETTh2

192 0.372 0.392 0.431 0.446 0.422 0.425 0.395 0.404 0.317 0.402 0.401 0.414 0.404 0.426 0.344 0.397
336 0.386 0.414 0.486 0.474 0.462 0.460 0.428 0.434 0.332 0.396 0.437 0.448 0.406 0.434 0.360 0.400
720 0.412 0.434 0.517 0.522 0.476 0.475 0.443 0.451 0.342 0.408 0.469 0.471 0.448 0.463 0.428 0.461
Avg 0.364 0.395 0.451 0.460 0.426 0.435 0.394 0.412 0.316 0.384 0.412 0.432 0.397 0.422 0.350 0.399
96 0.320 0.357 0.383 0.408 0.344 0.378 0.331 0.378 0.291 0.340 0.338 0.375 0.316 0.362 0.288 0.336
ETTm1

192 0.361 0.381 0.417 0.421 0.397 0.408 0.386 0.399 0.327 0.365 0.392 0.406 0.374 0.391 0.332 0.374
336 0.390 0.404 0.437 0.448 0.429 0.430 0.426 0.421 0.360 0.381 0.410 0.426 0.408 0.411 0.358 0.381
720 0.454 0.441 0.512 0.481 0.489 0.460 0.489 0.465 0.415 0.417 0.481 0.476 0.472 0.454 0.420 0.417
Avg 0.381 0.395 0.438 0.440 0.415 0.419 0.408 0.416 0.348 0.375 0.406 0.421 0.393 0.405 0.350 0.377
96 0.175 0.258 0.201 0.297 0.191 0.278 0.179 0.282 0.164 0.254 0.192 0.274 0.187 0.268 0.160 0.249
ETTm2

192 0.237 0.299 0.261 0.324 0.258 0.320 0.244 0.305 0.223 0.295 0.248 0.322 0.237 0.301 0.228 0.299
336 0.298 0.340 0.328 0.366 0.319 0.357 0.320 0.357 0.279 0.330 0.301 0.348 0.299 0.352 0.269 0.328
720 0.391 0.396 0.424 0.417 0.417 0.412 0.419 0.432 0.359 0.383 0.411 0.398 0.413 0.419 0.421 0.426
Avg 0.275 0.323 0.303 0.351 0.296 0.342 0.290 0.344 0.256 0.315 0.288 0.336 0.284 0.335 0.270 0.326

23
Published as a conference paper at ICLR 2024

Table 24: Short-term forecasting results in the PEMS datasets with multiple variates. All input lengths
are 96 and prediction lengths are 12. A lower MAE, MAPE or RMSE indicates a better prediction.
TimeMixer Scaleformer MTSMixer TSMixer
Models
(Ours) (2023) (2023) (2023)
MAE 14.63 17.66 18.63 15.71
PEMS03 MAPE 14.54 17.58 19.35 15.28
RMSE 23.28 27.51 28.85 25.88
MAE 19.21 22.68 25.57 20.86
PEMS04 MAPE 12.53 14.81 17.79 12.97
RMSE 30.92 35.61 39.92 32.68
MAE 20.57 27.62 25.69 22.97
PEMS07 MAPE 8.62 12.68 11.57 9.93
RMSE 33.59 42.27 39.82 35.68
MAE 15.22 20.74 24.22 18.79
PEMS08 MAPE 9.67 12.81 14.98 10.69
RMSE 24.26 32.77 37.21 26.74

Table 25: Short-term forecasting results in the M4 dataset with a single variate. All prediction lengths
are in [6, 48]. A lower SMAPE, MASE or OWA indicates a better prediction.
TimeMixer Scaleformer MTSMixer TSMixer
Models
(Ours) (2023) (2023) (2023)
SMAPE 13.206 13.778 20.071 19.845
Yearly MASE 2.916 3.176 4.537 4.439
OWA 0.776 0.871 1.185 1.166
SMAPE 9.996 10.727 16.371 16.322
Quarterly MASE 1.166 1.291 2.216 2.21
OWA 0.825 0.954 1.551 1.543
SMAPE 12.605 13.378 18.947 19.248
Monthly MASE 0.919 1.104 1.725 1.774
OWA 0.869 0.972 1.468 1.501
SMAPE 4.564 4.972 7.493 7.494
Others MASE 3.115 3.311 5.457 5.463
OWA 0.982 1.112 1.649 1.651
Weighted SMAPE 11.723 12.978 18.041 18.095
Average MASE 1.559 1.764 2.677 2.674
OWA 0.840 0.921 1.364 1.336

24
Published as a conference paper at ICLR 2024

J L IMITATIONS AND F UTURE W ORK

TimeMixer has shown favorable efficiency in GPU memory and running time as we presented in the
main text. However, it should be noted that as the input length increases, the linear mixing layer may
result in a larger number of model parameters, which is inefficient for mobile applications. To address
this issue and improve TimeMixer’s parameter efficiency, we plan to investigate alternative mixing
designs, such as attention-based or CNN-based in future research. In addition, we only focus on the
temporal dimension mixing in this paper, and also plan to incorporate the variate dimension mixing
into model design in our future work. Furthermore, as an application-oriented model, we made a
great effort to verify the effectiveness of our design with experiments and ablations. The theoretical
analysis to verify the optimality and completeness of our design is also a promising direction.

(a) TimeMixer (b) PatchTST (c) TimesNet (d) DLinear

(e) Autoformer (f) Scaleformer (g) MTSMixer (h) TSMixer

Figure 9: Prediction cases from ETTh1 by different models under the input-96-predict-96 settings.
Blue lines are the ground truths and orange lines are the model predictions.

(a) TimeMixer (b) PatchTST (c) TimesNet (d) DLinear

(e) Autoformer (f) Scaleformer (g) MTSMixer (h) TSMixer

Figure 10: Prediction cases from Electricity by different models under input-96-predict-96 settings.

25
Published as a conference paper at ICLR 2024

(a) TimeMixer (b) PatchTST (c) TimesNet (d) DLinear

(e) Autoformer (f) Scaleformer (g) MTSMixer (h) TSMixer

Figure 11: Prediction cases from Traffic by different models under the input-96-predict-96 settings.

(a) TimeMixer (b) PatchTST (c) TimesNet (d) DLinear

(e) Autoformer (f) Scaleformer (g) MTSMixer (h) TSMixer

Figure 12: Prediction cases from Weather by different models under the input-96-predict-96 settings.

Figure 13: Showcases from Solar-Energy by different models under the input-96-predict-96 settings.

26
Published as a conference paper at ICLR 2024

(a)TimeMixer (b)SCINet (c) TimesNet (d) Crossformer

(e) Autoformer (f) Scaleformer (g) MTSMixer (h) TSMixer

Figure 14: Showcases from PEMS03 by different models under the input-96-predict-12 settings.

(a)TimeMixer (b)SCINet (c) TimesNet (d) Crossformer

(e) Autoformer (f) Scaleformer (g) MTSMixer (h) TSMixer

Figure 15: Showcases from PEMS04 by different models under the input-96-predict-12 settings.

(a)TimeMixer (b)SCINet (c) TimesNet (d) Crossformer

(e) Autoformer (f) Scaleformer (g) MTSMixer (h) TSMixer

Figure 16: Showcases from PEMS07 by different models under the input-96-predict-12 settings.

27
Published as a conference paper at ICLR 2024

(a)TimeMixer (b)SCINet (c) TimesNet (d) Crossformer

(e) Autoformer (f) Scaleformer (g) MTSMixer (h) TSMixer

Figure 17: Showcases from PEMS08 by different models under the input-96-predict-12 settings.

(a)TimeMixer (b)SCINet (c) TimesNet (d) PatchTST

(e) Autoformer (f) Scaleformer (g) MTSMixer (h) TSMixer

Figure 18: Showcases from the M4 dataset by different models under the input-36-predict-18 settings.

28