№ |
Lecture topics |
Content |
Section 1. Introduction to econometrics |
1.1 |
Theoretical aspects of econometric modeling |
Econometrics as a science. The subject, method, and main tasks of econometrics. Sequence of econometric studies. |
1.2 |
Laws of distribution of random variables. Statistical conclusions: estimates and hypothesis testing |
Normal distribution. Distribution
. Student distribution. Fischer distribution. Distribution tables and their application. Point estimates and their properties. Interval estimates. Plotting confidence intervals. Errors of the 1st and 2nd kind. Criteria for testing statistical hypotheses. |
Section 2. Regression models |
2.1 |
Paired, multiple linear regression. Nonlinear regression |
Relationships of economic variables. The essence of regression analysis. The reasons for the presence of a random factor (deviation) in regression models. Paired linear regression. Multiple linear regression. The least squares method. Determining the parameters of the regression equation. Logarithmic (log-linear) models. Semi-logarithmic models. Reverse models. Power models. Illustrative models. Model selection. Specification issues. |
2.2 |
Checking the overall quality of the regression equation |
Prerequisites of the least squares method. Analysis of the accuracy of determining estimates of regression coefficients. Testing hypotheses about the coefficients of the regression equation. Coefficient of determination
. |
Section 3. Background of the least squares method |
3.1 |
Heteroscedasticity |
The essence of heteroskedasticity. Consequences of heteroskedasticity. Methods for detecting heteroscedasticity (graphical analysis of residuals, Spearman rank correlation test, Park test, Glazer test, Goldfeld-Quadt test). Methods for mitigating heteroscedasticity. |
3.2 |
Autocorrelation |
The essence and causes of autocorrelation. Consequences of autocorrelation. Methods for detecting autocorrelation (graphical analysis of residuals, series method, Darbin-Watson test). Methods for eliminating autocorrelation (Cochrane-Orcutt method, Hildreth-Lou method, first difference method). |
3.3 |
Multicollinearity |
The essence of multicollinearity. Consequences of multicollinearity. definition of multicollinearity. Methods for eliminating multicollinearity (excluding variables, getting a new sample, changing the model specification, converting variables). |