lab10 - Stat 500 lab notes Week 10 Autocorrelated errors c...

This preview shows pages 1–4. Sign up to view the full content.

Stat 500 lab notes c ± Philip M. Dixon, 2009 Week 10: Autocorrelated errors This week, I have done one possible analysis and provided lots of output for you to consider. Case study: predicting body fat Body fat is an important health measure, but accurately measuring body fat is not easy. The best method requires weighing someone underwater. A quicker, easier method, based on physical measurements, would be desirable. Bodyfat.txt on the class web site includes data on bodyfat and physical mea- surements for 252 men. A multiple regression using all variables was ﬁt. The regression output, diagnostic tables and some plots are included on subsequent pages. Some of the numerical output has been condensed. Questions for discussion: 1. Is this a good model? 2. How well does this model predict body fat? 3. What, if anything, concerns you? 4. What, if anything, would you do next? - 106 -

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Stat 500 lab notes c ± Philip M. Dixon, 2009 The SAS System 1 The REG Procedure Model: MODEL1 Dependent Variable: fat Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model 13 13168 1012.88783 54.65 <.0001 Error 238 4411.44804 18.53550 Corrected Total 251 17579 Root MSE 4.30529 R-Square 0.7490 Dependent Mean 19.15079 Adj R-Sq 0.7353 Coeff Var 22.48098 Parameter Estimates Parameter Standard Variance Variable DF Estimate Error t Value Pr > |t| Inflation Intercept 1 -18.18849 17.34857 -1.05 0.2955 0 age 1 0.06208 0.03235 1.92 0.0562 2.25045 weight 1 -0.08844 0.05353 -1.65 0.0998 33.50932 height 1 -0.06959 0.09601 -0.72 0.4693 1.67459 neck 1 -0.47060 0.23247 -2.02 0.0440 4.32446 chest 1 -0.02386 0.09915 -0.24 0.8100 9.46088 abdomen 1 0.95477 0.08645 11.04 <.0001 11.76707 hip 1 -0.20754 0.14591 -1.42 0.1562 14.79652 thigh 1 0.23610 0.14436 1.64 0.1033 7.77786 knee 1 0.01528 0.24198 0.06 0.9497 4.61215 - 107 -
Stat 500 lab notes c ± Philip M. Dixon, 2009 ankle 1 0.17400 0.22147 0.79 0.4329 1.90796 biceps 1 0.18160 0.17113 1.06 0.2897 3.61974 forearm 1 0.45202 0.19913 2.27 0.0241 2.19249 wrist 1 -1.62064 0.53495 -3.03 0.0027 3.37751 The REG Procedure Model: MODEL1 Dependent Variable: fat Output Statistics Hat Diag Cov ------DFBETAS----- Obs Residual RStudent H Ratio DFFITS Intercept age (output condensed) 30 -2.8265 -0.6703 0.0429 1.0792 -0.1418 0.0249 0.0180 31 -2.7030 -0.7546 0.3090 1.4844 -0.5046 0.1079 0.0263 32 -5.3312 -1.2775 0.0580 1.0228 -0.3168 0.1600 0.1313 33 5.6649 1.3645 0.0668 1.0187 0.3649 0.0838 -0.0503 34 -2.5185 -0.6005 0.0534 1.0970 -0.1427 0.0154 -0.0134 35 0.0297 0.007005 0.0312 1.0948 0.0013 -0.0001 0.0000 36 2.5204 0.6538 0.2001 1.2930 0.3270 -0.1603 -0.0191 37 0.0800 0.0188 0.0234 1.0862 0.0029 0.0003 -0.0007 38 6.4966 1.5538 0.0513 0.9701 0.3613 -0.1469 0.0347 39 -8.8349 -2.6280 0.3751 1.1354 -2.0362 -0.3904 -0.2800 40 0.0747 0.0178 0.0580 1.1260 0.0044 0.0003 -0.0011 41 -2.1759 -0.5693 0.2140 1.3239 -0.2970 0.0650 0.0125 42 0.3652 0.1660 0.7400 4.0735 0.2801 0.1325 -0.0015 43 -2.6349 -0.6337 0.0697 1.1134 -0.1734 0.0404 -0.0344

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 19

lab10 - Stat 500 lab notes Week 10 Autocorrelated errors c...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online