Download Econometrics Cheat Sheet and more Cheat Sheet Introduction to Econometrics in PDF only on Docsity! Applied Econometrics/Econometrics Mid-term Exam - Reference Sheet Simple linear regression (SLR) model Population model: ๐ฆ = ๐ฝ0 + ๐ฝ1๐ฅ + ๐ข Population model for a particular observation ๐ from a random sample with ๐ observations: ๐ฆ๐ = ๐ฝ0 + ๐ฝ1๐ฅ๐ + ๐ข๐, ๐ = 1, 2, . . , ๐ The dependent variable expressed in terms of estimates ?ฬ?โฒ๐ and ?ฬ? for observation ๐: ๐ฆ๐ = ?ฬ?0 + ?ฬ?1๐ฅ๐ + ?ฬ?๐ Regression residuals: ?ฬ?๐ = ๐ฆ๐ โ ?ฬ?๐ The Ordinary Least Squares (OLS) Estimators for SLR model: ?ฬ?1 = โ (๐ฆ๐โ?ฬ
?)(๐ฅ๐โ?ฬ
?) ๐ ๐=1 โ (๐ฅ๐โ?ฬ
?) 2๐ ๐=1 , ?ฬ?0 = ?ฬ
? โ ?ฬ?1?ฬ
? Total Sum of Squares Explained Sum of Squares Residual Sum of Squares SST โก โ (๐ฆ๐ โ ?ฬ
?) 2๐ ๐=1 SSE โก โ (?ฬ?๐ โ ?ฬ
?) 2๐ ๐=1 SSR โก โ ?ฬ?๐ 2๐ ๐=1 SST = SSE + SSR R-squared of the regression: ๐
2 โก SSE SST = 1 โ SSR SST Assumptions for the SLR model Assumption SLR.1 (Linear in parameters): In the population, the relationship between ๐ฆ and ๐ฅ is linear. ๐ฆ = ๐ฝ0 + ๐ฝ1๐ฅ + ๐ข Assumption SLR.2 (Random sampling): In a random sample with ๐ observations {(๐ฅ๐, ๐ฆ๐): ๐ = 1, 2, โฆ , ๐}, each data point follows the population equation. ๐ฆ๐ = ๐ฝ0 + ๐ฝ1๐ฅ๐ + ๐ข๐ Assumption SLR.3 (Sample variation in explanatory variable): The values of ๐ฅ๐ are not exactly the same. โ (๐ฅ๐ โ ?ฬ
?) 2 > 0๐๐=1 Assumption SLR.4 (Zero conditional mean): E(๐ข๐|๐ฅ๐) = 0 Assumption SLR.5 (Homoskedasticity): Var(๐ข|๐ฅ) = ๐2 The unbiased estimator of ๐๐ for SLR model: ?ฬ?2 = โ ๐ข๐ 2๐ ๐=1 ๐โ2 = SSR ๐โ2 Interpretation of ๐ท๐ for the functional forms involving logarithms: Model y x Interpretation of ๐ท๐ Model y x Interpretation of ๐ท๐ Level-level ๐ฆ ๐ฅ โ๐ฆ โ๐ฅ = ๐ฝ1 Log-level log ๐ฆ ๐ฅ %โ๐ฆ โ๐ฅ = (100๐ฝ1) Level-log ๐ฆ log ๐ฅ โ๐ฆ %โ๐ฅ = ( ๐ฝ1 100 ) Log-log log ๐ฆ log ๐ฅ %โ๐ฆ %โ๐ฅ = ๐ฝ1 Multiple linear regression (MLR) Model Population model: ๐ฆ = ๐ฝ0 + ๐ฝ1๐ฅ1 + ๐ฝ2๐ฅ2 + โฏ + ๐ฝ๐๐ฅ๐ + ๐ข R-squared of the regression: ๐
2 โก SSE SST = 1 โ SSR SST = (โ (๐ฆ๐โ?ฬ
?)(?ฬ?๐โ?ฬ
ฬ?)) ๐ ๐=1 2 (โ (๐ฆ๐โ?ฬ
?) 2)๐๐=1 (โ (?ฬ?๐โ?ฬ
ฬ?) 2 )๐๐=1