Econometric Analysis

Economic phenomena are quantified using economic theory, mathematics, and statistical inference.

What makes economic analysis crucial?

Economists can translate economic ideas into quantifiable metrics using econometrics. Econometrics is essential for identifying trends in different datasets. Economists can also predict upcoming monetary or economic trends based on current trends. Additionally, it aids in extracting a specific pattern or outcome from jumbled data.

Econometrics stages

• Stage 1: Create a hypothesis or theory. To direct data analysis, econometricians first develop a hypothesis or theory.
• Stage 2: Name your statistical model. Economists choose a statistical model in this step to look at the relationship between variables.
• Stage 3: Calculate the variables in the model.
• Stage 4: Conduct a test.

Econometrics data types

• Type 1: cross-sectional data
• Type 2: time-series data
• Type 3: pooled cross-sectional data
• Type 4: longitudinal (aka panel) data

1. Basics of probability and statistics

1.1 Random variables and probability distributions

1.1.1 Properties of probabilities

1.1.2 The probability function – the discrete case

1.1.3 The cumulative probability function – the discrete case

1.1.4 The probability function – the continuous case

1.1.5 The cumulative probability function – the continuous case

1.2 The multivariate probability distribution function

1.3 Characteristics of probability distributions

1.3.1 Measures of central tendency

1.3.2 Measures of dispersion

1.3.3 Measures of linear relationship

1.3.4 Skewness and kurtosis

2. Basic probability distributions in econometrics

2.1 The normal distribution

2.2 The t-distribution

2.3 The Chi-square distribution

2.4 The F-distribution

3. The simple regression model

3.1 The population regression model

3.1.1 The economic model

3.1.2 The econometric model

3.1.3 The assumptions of the simple regression model

3.2 Estimation of population parameters

4. Statistical inference

4.1 Hypothesis testing

4.2 Confidence interval

4.2.1 P-value in hypothesis testing

4.3 Type I and type II errors

4.4 The best linear predictor

5. Model measures

5.1 The coefficient of determination (R2)

5.2 The adjusted coefficient of determination (Adjusted R2)

5.3 The analysis of variance table (ANOVA)

6. The multiple regression model

6.1 Partial marginal effects

6.2 Estimation of partial regression coefficients

6.3 The joint hypothesis test

6.3.1 Testing a subset of coefficients

6.3.2 Testing the regression equation

7. Specification

7.1 Choosing the functional form

7.1.1 The linear specification

7.1.2 The log-linear specification

7.1.3 The linear-log specification

7.1.4 The log-log specification

7.2 Omission of a relevant variable

7.3 Inclusion of an irrelevant variable

7.4 Measurement errors

8. Dummy variables

8.1 Intercept dummy variables

8.2 Slope dummy variables

8.3 Qualitative variables with several categories

8.4 Piecewise linear regression

8.5 Test for structural differences

9. Heteroskedasticity and diagnostics

9.1 Consequences of using OLS

9.2 Detecting heteroskedasticity

9.2.1 Graphical methods

9.2.2 Statistical tests

9.3 Remedial measures

9.3.1 Heteroskedasticity-robust standard errors

10. Autocorrelation and diagnostics

10.1 Definition and the nature of autocorrelation

10.2 Consequences

10.3 Detection of autocorrelation

10.3.1 The Durbin Watson test

10.3.2 The Durbins h test statistic

10.3.3 The LM-test

10.4 Remedial measures

10.4.1 GLS with AR(1)

10.4.2 GLS with AR(2)