3_multiple_regression

# 3_multiple_regression - THE MULTIPLE REGRESSION MODEL A...

This preview shows pages 1–3. Sign up to view the full content.

RESEARCH METHODS IN ECONOMICS ECO 4451 Slide # 1 THE MULTIPLE REGRESSION MODEL: A REVIEW y = β 0 + 1 x 1 + 2 x 2 + 3 x 3 + ε Slide # 2 I. INTRODUCTION ´ A. Definition « 1. Previously: simple two-variable model « 2. Now: more realistic « 3 Multiple regression mode Slide # 3 3. Multiple regression model « 4. Multiple IVs (X 1 , X 2 , . . . X k ) INTRODUCTION (CONT.) ´ B. The Model « 1. Y = β 0 X 0 + β 1 X 1 + β 2 X 2 + . . . + β k X k + ε « 2. NOTE: ² a β is intercep Slide # 4 a) 0 is intercept ² b) X 0 is constant term (column of 1's ) « 3. Y = β 0 + β 1 X 1 + β 2 X 2 + . . . + β k X k + ε « 4. coefficients to estimate ² Excluding intercept , K ² Including intercept , K+1 X 0 IS CONSTANT TERM (COLUMN OF 1'S) PROFIT Constant MKTSIZE WINS 33.4 1 3 57 22 1 3 63 16 1 3 46 8.7 1 2 42 Slide # 5 5.4 1 2 44 4.7 1 2 55 -1.5 1 1 35 -2.1 1 1 13 -4 1 1 28 NOTE: NBA market size = 3 for large, 2 for medium, 1 for small PROFIT is in millions of \$ WHO USES THIS The fuel economy of engine designs can be predicted by multiple regression models . The Building Owner Management Association of Chicago uses a multiple regression model to predict Slide # 6 downtown commercial rental rates. Their report not only presents the multiple regression coefficients but also the t-ratio of each coefficient. (Source: Professor John McDonald Department of Economics, University of Illinois at Chicago.)

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
WHO USES THIS COLORADO STATE HURRICANE FORECAST TEAM SEES INCREASED HURRICANE ACTIVITY Colorado State Press Release: May 23, 2002 FORT COLLINS-After six years that set a record for Atlantic Basin tropical storms and hurricane activity William Gray and Slide # 7 Basin tropical storms and hurricane activity, William Gray and his associates at Colorado State University are calling for a season slightly above the long-term average. For 2002, they forecast twelve named storms, seven hurricanes and three major (Saffir-Simpson category 3-4-5) hurricanes. Gray and co-authors . . . apply statistical regression . . . to make their forecasts. Remember summer of 2004?? INTRODUCTION (CONT.) ´ C. Interpreting each β k Y = β 0 + β 1 X 1 + β 2 X 2 + . . . + β k X k + ε Slide # 9 ´ When X k changes by one of its units, Y changes by β k (in Y's units). . . holding all other Xs fixed INTERPRETING EACH Β K (CONT.) PROFIT = β 0 + β 1 MKTSIZE + β 2 WINS + ε « PROFIT = NBA team annual profit (\$M), « MKTSIZE= market size (1=small, 2=medium, 3= large), « WINS= number of wins in one season Slide # 10 PROFIT = -20.8 + .185 WINS + 11.2 + .185 WINS + 11.2 MKTSIZE MKTSIZE ´ As market size increases by one category, profit rises \$11,200,000, even if number of wins stay the same. (average rise) INTERPRETING EACH Β K (CONT.) PROFIT = β 0 + β 1 MKTSIZE + β 2 WINS + « PROFIT = NBA team annual profit (\$M), « MKTSIZE= market size (1=small, 2=medium, 3= large), « WINS= number of wins in one season Slide # 11 PROFIT = -20.8 + .185 WINS + 11.2 + .185 WINS + 11.2 MKTSIZE MKTSIZE ´ As team wins one more game, profit rises \$????, even if ???? stays the same.
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 15

3_multiple_regression - THE MULTIPLE REGRESSION MODEL A...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online