数理统计——多元回归.pptVIP

Less chance of error than separate t-tests on each coefficient. Doing a series of t-tests leads to a higher overall Type I error than ?. As a result of this class, you will be able to... As a result of this class, you will be able to... Is this model specified correctly? What other variables could be used (color, photo’s etc.)? As a result of this class, you will be able to... Note potential problem with multicollinearity. This is solved somewhat by centering on the mean. Multiple Regression Evaluate Your Multiple Regression Model by Using: Use Adjusted R-Square Use Statistical Tests Analysis of Residual Multicollinearity Collinearity 1. High Correlation Between X Variables 2. Coefficients Measure Combined Effect 3. Leads to Unstable Coefficients Depending on X Variables in Model 4. Always Exists - Matter of Degree 5. Example - Using Both Sales Profit as “Independent” Variables in Same Model Detecting Collinearity 1. Examine Correlation Matrix Correlations Between Pairs of X Variables (‘Independent Variables’) Are More than With Y Variable Or if correlation between our ‘Independent Variables’ is more than .80), then Multicollinearity Exists (your book uses 0.90) 2. Few Remedies Obtain New Sample Data Eliminate One Correlated X Variable Example (NYT) You work in advertising for the New York Times. You want to find the effect of ad size (sq. in.) newspaper circulation (in thousands of copies) on the number of ad responses (in hundreds of responses). You’ve collected the following data: Resp Size Circ 1 1 2 4 8 8 1 3 1 3 5 7 2 6 4 4 10 6 Correlation Matrix Computer Output Explain the Non-Linear Regression Model Polynomial (Curvilinear) Regression Model 1. Relationship Between 1 Dependent 2 or More Independent Variables Is a Quadratic Function 2. Useful 1st Model If Non-Linear Relationship Suspected 3. Polynomial Model Linear Effect Curvilinear Effect Polynomial (Curvilinear) Model Relationships (did not cover this slide but you may know it) Y X 1 Y