MKT 317 Lecture Notes - Lecture 13: Simple Linear Regression, Explained Variation, F-Test
Document Summary
Observed value of yi minus mean of yi of dependent variable. Predicted value of yi minus the mean of yi. Unexplained variation sse= (cid:4666)yiy i(cid:4667)(cid:2870)= (cid:2924)i=(cid:2869) (cid:2904)(cid:2925)ta(cid:2922) variati(cid:2925)(cid:2924) = (cid:2903)(cid:2903)(cid:2902)(cid:2903)(cid:2903)(cid:2904) Observed value of yi minus predicted value of yi. (cid:4666)ei(cid:4667)(cid:2870) (cid:2924)i=(cid:2869: 0 r2 1; usually expressed as a percentage, closer r2 is to 1, the better the model since explains more of the variation in the data. H1: 1 0 f-test for regression. Dfnum = k = 1 for simple linear regression (only have 1 independent variable) K = # of independent variables; in simple regression k = (cid:883) Coefficient determination r2/ 1 variable already explaining the variation. R2 = (cid:2903)(cid:2903)(cid:2902)(cid:2903)(cid:2903)(cid:2904)= (cid:2876)(cid:2875)(cid:2877). (cid:2871)5 (cid:2902)(cid:2870) (cid:4666)(cid:2869) (cid:2902)2(cid:4667)(cid:4666)(cid:2924) (cid:2870)(cid:4667) estimate which in this case is b0 & b1. The unemployment rate explains 90. 5% of the variation in the # of grievances filled. F test of overall model ( = 0. 05. H1: 1 0 k = 1, n = 7.