Econometrics: Study Guide
1. What is Econometrics?
Econometrics is a branch of economics that applies statistical and mathematical methods to
economic data. It tests hypotheses, estimates economic relationships, and forecasts trends by
combining economic theory, mathematics, and statistics to provide quantitative insights.
Purpose of Econometrics
1. Estimate economic models (e.g., demand and supply functions).
2. Test economic theories.
3. Forecast future trends (e.g., GDP growth, inflation).
4. Evaluate policies (e.g., effect of minimum wages on employment).
2. Types of Econometrics
Econometrics is broadly divided into:
A. Descriptive Econometrics - Summarizes data using statistical measures like mean, median, and
variance.
Example: Analyzing average income or unemployment trends.
B. Theoretical Econometrics - Develops statistical methods and assumptions for economic models.
Example: Classical Linear Regression Model (CLRM) assumptions.
C. Applied Econometrics - Uses real-world data to test and estimate economic models.
Example: Estimating the relationship between income (Y) and consumption (C) using regression
models.
3. Core Econometric Models with Formulas and Examples
(i) Simple Linear Regression:
Y = b0 + b1X + u
Example: Salary vs. years of experience.
If Y = 20000 + 1500X, and X = 5 years, then Y = 27,500.
(ii) Multiple Linear Regression:
Y = b0 + b1X1 + b2X2 + ... + bkXk + u
Example: Predicting house prices based on size and rooms.
Price = 50000 + 300(Size) + 2000(Rooms).
(iii) Hypothesis Testing:
Null Hypothesis H0: b1 = 0
t = (b1_hat - b1)/SE(b1_hat)
(iv) Time Series Models:
AR Model: Yt = phi0 + phi1Yt-1 + ut.
Example: Forecasting inflation.
(v) Panel Data Models:
Yit = ai + bXit + uit.
4. Real-World Example
Suppose we study how education affects wages:
Wage = b0 + b1 Education + u
If Wage = 5000 + 200 * Education,
then each additional year of schooling increases wages by 200 units.
5. Key Steps in Econometrics
1. Specify the model.
2. Collect data (cross-sectional, time-series, or panel).
3. Estimate parameters (using OLS, MLE, etc.).
4. Test hypotheses (using t-test, F-test).
5. Forecast or interpret results.