Econometrics

University of Milan-Bicocca

Course lecturer:
Maryam Ahmadi
[email protected]

1
Introduction: The Nature of
Econometrics
and Economic Data

2
The Nature of Econometrics and Economic Data
• What is econometrics?
• Econometrics = use of statistical methods to analyze economic data

• Typical goals of econometric analysis

• Development of statistical methods for estimating relationships between
variables
For example between individual wages and the level of schooling
• Testing economic theories and hypotheses
• Forecasting economic variables (such as interest rates, inflation, gross
domestic product , …)
• Evaluating and implementing government and business policies
3
Steps in an econometric analysis

1. Constructing a formal economic model

Example: Model of job training and worker productivity

4
Steps in an econometric analysis
• Example
• What is the effect of additional training on worker productivity?

Hourly wage Weeks spent

Years of formal Years of workforce in job training
education experience

• Other factors may be relevant, but these are the most important ones
• the choice of these variables is determined by the economic theory as well as
data considerations.
5
 Steps in an econometric analysis

2. Econometric model

Econometric model of job training and worker productivity

Hourly wage Years of formal Years of workforce Weeks spent Unobserved determinants of the wage
education experience in job training
e.g. innate ability, quality of education,
family background …

6
 Steps in an econometric analysis

• The u term contains the unobserved factors and is called the error term or disturbance term.
Dealing with this error term is perhaps the most important component of any econometric
analysis. It controls for the effects of other variables on wage and therefore excludes the
effects on unobserved variables from the model parameters.

• β0 , β1 , β2 , β3 are the parameters of our econometric model. The parameters describe the
directions and strengths of the relationship between wage and the factors used to determine
wage in the model.

• Econometric models may be used for hypothesis testing

• For example, the parameter β3 represents “effect of training on wage”
• How large is this effect? Is it different from zero?
7
Steps in an econometric analysis

3. The structure of data

• Econometric analysis requires data
• Econometric methods depend on the nature of the data used
• Use of inappropriate methods may lead to misleading results

• There are different kinds of data sets

• Cross-sectional data
• Time series data
• Panel data

8
 Steps in an econometric analysis

• Cross-sectional data sets

• Sample of individuals, households, firms, cities, states, countries, or

other units of interest at a given point of time or in a given period

• An important feature of cross-sectional data is that they have been

obtained by random sampling from a population.

Example: obtaining information on wages, education, experience,

… by randomly drawing N people from the population all working
people.

9
Steps in an econometric analysis

• Cross-sectional data sets

• Cross-sectional data typically encountered in applied
microeconomics, such as labour economics, state and local public
finance, industrial organization, urban economics, and health
economics.

• Data on individuals, households, firms, and cities at a given point in

time are important for testing microeconomic hypotheses and
evaluating economic policies.

❑In cross sectional data sets ordering of observations does not

matter.

10
Steps in an econometric analysis

• Cross-sectional data set on wages and other characteristics for 526 working
individals for the year 1976. (WAGE1 Stata)

Indicator variables
Dummy variables
(1 = yes, 0 = no)

Observation number Years of Years of

Hourly wage
education experience
11
 Steps in an econometric analysis
• Time series data
• Time series data set consists of observations of variables over time.
Example: stock prices, money supply, consumer price index, gross domestic
product, automobile sales, …

❑Past events influence future events, therefore, ordering of observations

has important information.

• Time series observations are typically serially correlated (the

observations are correlated across time, GDP example).

• Serial correlation makes time series data more difficult to analyze than
cross-sectional data. Therefore, more needs to be done in specifying
econometric models for time series data before applying a standard
econometrics method.

12
 Steps in an econometric analysis
• Time series data

• The most common data frequencies are: daily, weekly, monthly, quarterly,
annually, …

• For example, stock prices are reported at daily intervals, the money supply in the US
economy is reported weekly, many macroeconomic variables such as inflation and
unemployment rates are reported at monthly frequency, GDP normally is reported quarterly,
and some time series such as mortality rates are available only on an annual basis.

• Different observations refer to different times and it does not make sense
to assume random sample of time periods.

• Typical applications: applied macroeconomics and finance

13
Steps in an econometric analysis

• Panel data
• Panel data set consists of a time series for each cross-sectional
member in the data set.
• Cross-sectional units are followed over time
Example: data on wage and education of sample of 500 workers, over a ten-
year period.
For each worker we have 2 time series, that is his wage over ten years and his
level of education over ten years.
For each time, we have 2 cross sectional data sets: wage of 500 workers and
education level of 500 workers.
• Panel data have a cross-sectional and a time series dimension

14
Steps in an econometric analysis

4) Applying Econometrics methods on our dataset to

estimate the econometric model

• Ordinary Least Square (OLS):

a method to estimate unknown parameters of our econometrics model.

15
 Summary
Explaining y in terms of 𝑥1 , 𝑥2 , 𝑥3 ,…, 𝑥𝑘 “or” studying how y varies with changes in 𝑥1 , 𝑥2 , 𝑥3 ,…, 𝑥𝑘
The econometric mode
collect data for the model variables
Use the estimation method to estimate the unknown parameters of interest.

Intercept Slope parameters

Dependent variable,
explained variable, Error term,
Independent variables, disturbance,
response variable,…
explanatory variables, unobservables,…
regressors,…

16
Problem
All questions are to be answered True or False. If you answer True, then no explanation is needed to
justify your answer. If you answer False, then you need to state why it is False.

• a) An important job of econometrics is to quantify the relationships between different variables based on
• available data and using estimation methods, and to interpret and use the resulting outcomes appropriately.

• b) By estimation the following econometric model

The estimation result for quantifies the effect of worker’s training on his wage.

• c) By estimation the following econometric model

The estimation result for quantifies the effect of worker’s training on his wage.

Deadline: Monday 9th March, 2020, 24:00

17