# Question

The statistician George Box, who had an illustrious academic career at the University of Wisconsin, is often quoted as saying, “All models are wrong, but some models are useful.” Why do you think that, in practice,

a. All models are wrong?

b. Some models are not useful?

a. All models are wrong?

b. Some models are not useful?

## Answer to relevant Questions

In regression modeling, for t tests about regression parameters, df = n - number of parameters in equation for the mean. a. Explain why df = n - 2 for the model μy = α + βx. b. Chapter 8 discussed how to estimate a single ...A Freddie Mac quarterly statement (May 2010) reported that U.S. home sales for one of the central regions (including Illinois, Indiana, Ohio, and Wisconsin) have shown that home values decreased by 3.4% in the last previous ...For the Softball data set on the text CD, for each game the variables are a team’s number of runs scored (RUNS), number of hits (HIT), number of errors (ERR), and the difference (DIFF) between the number of runs scored by ...The P-value of 0.17 in part a of the previous exercise suggests that LP_200 plausibly had no effect on BP, once BP_60 is in the model. Yet when LP_200 is the sole predictor of BP, the correlation is 0.58 and the significance ...In Chapter 12, we analyzed strength data for a sample of female high school athletes. The following figure is a residual plot for the multiple regression model relating the maximum number of pounds the athlete could bench ...Post your question

0