STAT 2000 Lecture 8: January 30

• School: University of Georgia
• Department: Statistics
• Course: STAT 2000
• Professor: Dunlap

Document Summary

Sample 6: buying a used ford: 1. Association= linear; direction= negative; no outliers: 2. Strong association: 3. (cid:1877) =(cid:884)(cid:882)(cid:888)(cid:882)(cid:886) (cid:883)(cid:887)(cid:887)(cid:889). (cid:890)(cid:1876) price= intercept-slope(age, 4. (cid:1877) =(cid:884)(cid:882)(cid:888)(cid:882)(cid:886) (cid:883)(cid:887)(cid:887)(cid:889). (cid:890)(cid:4666)(cid:890)(cid:4667)=,(cid:883)(cid:886)(cid:883). (cid:888)(cid:882, 5. Newspaper price=,600; this is less than we expected to pay. If a price point falls above the line, it is a bad deal; below the line is a good deal: other variables that could predict the price: age, mileage, condition of car, number of previous owners, colors, mpg. Residual: the difference between the observed value of the response variable and the. 2= (cid:3028)(cid:3028)(cid:3042)(cid:3041) (cid:3032)(cid:3051)(cid:3043)(cid:3039)(cid:3028)(cid:3041)(cid:3032)(cid:3031) (cid:3029)(cid:3052) (cid:3040)(cid:3042)(cid:3031)(cid:3032)(cid:3039) (cid:3042)(cid:3028)(cid:3039) (cid:3028)(cid:3028)(cid:3042)(cid:3041) Unexplained variation (i. e. mileage, condition of car, etc: can be positive or negative, if point is above regression line; the point is a positive residual value predicted by regression line, residual=actual-predicted= (cid:1877) (cid:1877) . 2- squared correlation coefficient: 2= % of this box that variable (i. e. age) takes up total variation in responses=variation explained by model+ variation not explained.