Here's a concise and comprehensive note on Chapter 1 that will help you answer
20-mark questions for your exam on econometrics:
1. What is Econometrics?
• Econometrics is the application of statistical methods to test economic
theories, estimate economic relationships, and evaluate policies.
• It helps in forecasting economic variables (e.g., inflation, GDP) and
studying areas beyond macroeconomics, like political campaigns or educa-
tion spending.
2. Economic vs Econometric Models
• Economic Model: A theoretical framework using mathematical equations
to describe economic relationships (e.g., utility maximization).
• Econometric Model: A statistical representation of an economic model,
used to analyze real-world data. It involves the application of statistical
techniques like regression analysis.
3. Steps in Econometric Analysis
1. Formulate the Question: Define the issue, such as testing a theory or
analyzing policy impact.
2. Construct a Model: Based on economic theory, formulate an equa-
tion (e.g., wage = f(education, experience, training) for job train-
ing and productivity).
3. Translate to Econometric Model: Use statistical methods to quantify
relationships.
4. Estimate Parameters: Use data to estimate the parameters and test
hypotheses.
5. Interpret Results: Draw conclusions and make policy recommendations.
Model of Job Training and Worker Productivity
• Example: A model like wage = f(education, experience, training)
explores how job training affects wages.
• Unobserved Determinants of Wages:
– Example: Factors like innate ability, family background, and educa-
tion quality affect wages but are hard to measure directly.
– Solution: These are captured by the error term in the econometric
model (u), which represents unobserved variables.
It seems you've uploaded an image of an econometric model for job training and
worker productivity. Let's correct and incorporate the model into the earlier
notes:
1
Corrected Model of Job Training and Worker Productivity:
wage = β0 + β1 educ + β2 exper + β3 training + u
Where:
• wage = Hourly wage
• educ = Years of formal education
• exper = Years of workforce experience
• training = Weeks spent in job training
• u = Unobserved determinants of the wage (e.g., innate ability, quality of
education, family background)
This model demonstrates how education, experience, and training impact
wages, while u accounts for unobserved variables that could affect the wage but
are not directly measurable .
4. Structure of Economic Data & Types of Data Sets
1) Cross-sectional Data
• Definition: Data collected at a single point in time from various subjects
(e.g., individuals, firms).
• Example: Survey data on wages, education, and experience of individuals
in 1976.
• Features: Can assume random sampling; used in microeconomics, labor
economics, etc.
2) Time Series Data
• Definition: Data collected over time on a single subject (e.g., GDP,
inflation rates).
• Example: Monthly unemployment rates over a year.
• Features: Time order matters; special treatment needed for trends and
seasonality.
3) Pooled Cross-sectional Data
• Definition: Combining cross-sectional data from different periods to
increase sample size.
• Example: Housing price data from 1993 and 1995 to study tax effects.
• Features: Combines features of cross-sectional and time-series data.
4) Panel (Longitudinal) Data
• Definition: Data collected over time for the same subjects (e.g., individu-
als, firms).
2
• Example: Tracking the same individuals’ wages and education over 5
years.
• Features: Allows for the study of individual-level changes and more robust
causal inference.
Data Type Definition Example Features
Cross-sectional Snapshot at a Survey of wages Random
point in time in 1976 sampling, used in
microeconomics
Time Series Data over time on Monthly Time dependency,
a single subject unemployment requires special
rates treatment
Pooled Combined data Housing prices Increases sample
Cross-section from multiple before/after tax size, shows
cross sections temporal change
Panel Data Repeated data for Individual wages Allows
same subjects over years individual-level
over time analysis, causal
inference
5. Ceteris Paribus
• Definition: The assumption that all other relevant factors are held con-
stant while studying the effect of one variable.
• Application: Econometric analysis often uses this assumption to isolate
the effect of one factor (e.g., price) on another (e.g., quantity demanded).
6. Causality and Notion of Ceteris Paribus
• Causality: A relationship where a change in one variable (e.g., education)
leads to a change in another (e.g., wages). Econometrics aims to estimate
this relationship while holding other factors fixed (ceteris paribus).
• Example: In job training, we aim to understand how additional training
affects wages, while keeping education and experience constant.
These notes cover the core concepts, data types, and econometric models you'll
need to answer questions in your final exam.