MMMASCI CMM224

---Forecasting

Group 2(Wang Guangyu&Qiu Xiaoyu)

May 15, 2023

Contents:
1 What is forecasting?

2 Why we need to do forecasting ?

3 How to do forecasting ?

4 Example :XXX Airport Passenger Forecasting

5 Class Activity
1 What is forecasting?

Forecasting is the process of predicting or estimating future events, trends or outcomes

based on historical data and other relevant information. It is a critical aspect of many
fields, including finance, economics, business, and science, where the ability to make
informed decisions about the future is essential. The goal of forecasting is to provide
insights into what may happen in the future, enabling individuals and organizations to
make more accurate plans, reduce uncertainty, and make better decisions. Forecasting
techniques may involve statistical models, machine learning algorithms, time series
analysis, or other methods to identify patterns and trends in historical data that can be
used to predict future outcomes. However, it's important to keep in mind that forecasts are
not always precise or accurate, and there is always some level of uncertainty associated
with them.
2 Why we need to do forecasting ?

Forecasting is essential because it allows individuals and organizations to make informed

decisions about the future based on current and historical data. Here are some reasons why
we need forecasting:
①Planning: Forecasting helps individuals and organizations to plan for the future by anticipating trends and outcomes.
This can include forecasting demand for products or services, future sales revenue, or identifying potential risks and
opportunities.
②Resource allocation: Forecasting enables organizations to allocate resources effectively by predicting future demand
for resources, such as labor, raw materials, or inventory.
③Budgeting: Forecasting helps organizations to create accurate budgets by predicting future revenue and expenses.
This enables them to allocate funds effectively and make informed financial decisions.
④Risk management: Forecasting allows organizations to identify potential risks and develop strategies to mitigate or
avoid them. This can include predicting changes in the market, identifying potential supply chain disruptions, or
anticipating changes in customer behavior.
⑤Performance evaluation: Forecasting enables organizations to evaluate their performance by comparing actual
results with predicted outcomes. This can help them identify areas for improvement and make informed decisions
about future investments and initiatives.
Overall, forecasting is a critical tool for individuals and organizations across a wide range of
industries and fields, helping them to make more informed decisions, reduce uncertainty, and
achieve their goals more effectively.
3 How to do forecasting ?

Forecasting is the process of making predictions about future events based on historical data and
trends. Here are some general steps for doing forecasting:
①Define the problem: Clearly define what you want to forecast, such as sales, revenue, or customer demand.
②Gather data: Collect relevant data about the problem you want to forecast. The data should be representative of the past and
present trends in the problem.
③Clean and preprocess data: Ensure that the data is accurate, complete, and consistent. Remove any outliers, errors, or missing
values. Also, transform the data to make it suitable for modeling.
④Choose a forecasting model: There are various statistical and machine learning models available for forecasting. Choose a model
that is appropriate for your problem and data.
⑤Train the model: Use the historical data to train the forecasting model. The model should learn the patterns and relationships in the
data.
⑥Validate the model: Use a separate set of data (validation set) to evaluate the performance of the model. The validation set should
be representative of the data used for training the model.
⑦Make forecasts: Once the model is validated, use it to make forecasts on new data. Monitor the accuracy of the forecasts and
adjust the model if necessary.
⑧Communicate results: Communicate the forecasting results to stakeholders, including any uncertainties or assumptions made in
the forecasting process.
Remember that forecasting is not an exact science, and there is always uncertainty and risk involved.
Therefore, it is essential to use multiple models, assumptions, and scenarios to ensure that the
forecasting results are robust and reliable.
4 Example :XXX Airport Passenger Forecasting

Exponential Smoothing (ETS) algorithm

4 Example :XXX Airport Passenger Forecasting
5 Class Activity

1.In forecasting techniques, what are the common qualitative models?

2.In forecasting techniques, what are the common quantitative models?

3.Knowing the sales data of a certain company from 2017 to 2022, use the linear regression
method to predict the sales volume of the company in 2023/2024/2025.

Year Year(X) Sales(Y)

2017 1 11000
2018 2 12200
2019 3 12500
2020 4 15800
2021 5 15520
2022 6 16980
2023 7 ?????
2024 8 ?????
2025 9 ?????
THANK YOU FOR WATCHING