STA302H5 Lecture 18: Lecture 18 - 20131111.pdf
Document Summary
Used car sales, n = number of dealerships. > head(carsales) salesforce yrsexp yrseduc women income nsales avprice. > auto = lm(avprice ~ salesforce + yrsexp + yrseduc + women + income, data=carsales) Call: lm(formula = avprice ~ salesforce + yrsexp + yrseduc + women + income, data = carsales) Error t value pr(>|t|) (intercept) 6099. 1463 96. 8659 62. 965 < 2e-16 *** salesforce 1. 5577 1. 6865 0. 924 0. 35724 yrsexp 14. 4620 5. 3166 2. 720 0. 00733 ** yrseduc -13. 3943 7. 0424 -1. 902 0. 05918 . women 1. 0157 0. 6621 1. 534 0. 12721 income 0. 6383 0. 3099 2. 059 0. 04126 * Residual standard error: 98. 94 on 144 degrees of freedom. F-statistic: 3. 57 on 5 and 144 df, p-value: 0. 0045. > yhat = auto. values ; sr = rstandard(auto) > attach(carsales) # make variable names accessible without dollar signs. The following object(s) are masked from "package:datasets": women. > # one possible high outlier, one low. > sdr = rstudent(auto) # studentized deleted residuals.
