This preview shows pages 1–6. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Regression Analysis Harry Tsang Department of Economics University of Illinois, UrbanaChampaign April 11, 2008 Linear Regression: We Explain The Variation In Y, dependent variable, By Using Explanatory Variables X, independent variable I Simple Linear Regression  One Independent Variable X Y = + 1 X 1 + I Multiple Linear Regression  More Than One Independent Variable X Y = + 1 X 1 + 2 X 2 + ... + k X k + We Break The Total Variability In Y Into Two Different Components, SSR and SSE I SST (Sum of Squares Total): Total Amount of Variation, deg. of freedom = n  1 SST = n X i ( y i y ) 2 I SSR (Sum of Squares Regression): Total Amount of Variation Explained By Model, deg. of freedom = k (# of Independent Variables) SSR = n X i ( y i y ) 2 I SSE (Sum of Squares Error): Total Amount of Variation Unexplained By Model, deg. of freedom = n  k  1 SSE = n X i ( y i y i ) 2 How Good Is Our Model At Explaining the Variability In Y? I Coefficient of Determination, R 2 ,: What proportion of the variation in y is explained by our model? R 2 = SSR SST = 1 SSE SST I Ftest For Overall Model Validity, d.f. = k, nk1 H : 1 = 2 = ... = k = 0 H 1 : At least one is not equal to zero F = MSR MSE = SSR k SSE n k 1 I ttest For Significance Of Slope Coefficient, i , d.f. = n  k 1 H : i = 0 , H 1 : i 6 = 0 , H 1 : i < , H 1 : i > t = b i i s b i Multiple Linear Regression Y = + 1 X 1 + 2 X 2 + ... + k X k + I Standard Error of The Estimate s = r SSE n k 1 I Standard Error of The Slope Coefficient: s b i = Given in Excel output I Adjusted R 2 Adj R 2 = 1 SSE n k 1 SST n 1 = 1 SSE SST n...
View Full
Document
 Spring '08
 Petry
 Economics

Click to edit the document details