In the Department of Fisheries and Wildlife at Virginia Polytechnic Institute and State University, an experiment was conducted to study the effect of stream characteristics on fish biomass. The regressor variables are as follows: average depth (of 50 cells) (x1): area of in-stream cover (i.e., undercut banks, logs, boulders, etc.) (x2) percent canopy cover (average of 12) (x3) area > 25 centimeters in depth (x4). The response is y, the fish biomass. The data are as follows:(a) Fit a multiple linear repression including all four regression variables.(b) Use Cp, R2, and s to determine the best subset of variables. Compute these statistics for all possible subsets.(c) Compare the appropriateness of the models in parts (a) and (b) for predicting fishbiomass.

Transcribed Image Text:

Obs. 12.2 26.0 24.2 26.1 31.6 23.3 13.0 7.5 48.0 152.2 469.7 485.9 87.6 6.9 192.9 105.8 100 388 755 1288 230 14.3 19.1 54.6 28.8 16.1 10.0 28.5 13.8 15.0 29.4 58.0 42.6 15.9 56.4 551 95.1 345 60.6 Obs. X2 0.0 116.6 10.7 35.2 40.3 40.3 10 348 25.9 52.0