day8prlm

day8prlm - 3/16/12 1 PADP 8130: Linear Models Data Problems...

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

View Full Document Right Arrow Icon

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

View Full DocumentRight Arrow Icon

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

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 3/16/12 1 PADP 8130: Linear Models Data Problems PRACTICE Angela Fer¡g, Ph.D. Detec¡ng mul¡collinearity: VIF . collin educhd hsdropout collgrad somecoll Collinearity Diagnostics SQRT R- Variable VIF VIF Tolerance Squared---------------------------------------------------- educhd 5.90 2.43 0.1696 0.8304 hsdropout 2.17 1.47 0.4602 0.5398 collgrad 4.32 2.08 0.2315 0.7685 somecoll 1.89 1.38 0.5282 0.4718---------------------------------------------------- Mean VIF 3.57 . collin hsdropout hsgrad somecoll collgrad Collinearity Diagnostics SQRT R- Variable VIF VIF Tolerance Squared---------------------------------------------------- hsdropout 2.39e+13 4.9e+06 0.0000 1.0000 hsgrad 3.47e+13 5.9e+06 0.0000 1.0000 collgrad 2.81e+13 5.3e+06 0.0000 1.0000 somecoll 2.98e+13 5.5e+06 0.0000 1.0000---------------------------------------------------- Mean VIF 2.91e+13 VIF<10 so all fne VIF>10 so need to drop one o¢ these variables. This is why you have to have a re¢erence category. 3/16/12 2 Regression check for mul:collinearity reg hsgrad hsdropout collgrad somecoll Source | SS df MS Number of obs = 8277-------------+------------------------------ F( 3, 8273) = . Model | 1823.74435 3 607.914784 Prob > F = . Residual | 0 8273 0 R-squared = 1.0000-------------+------------------------------ Adj R-squared = 1.0000 Total | 1823.74435 8276 .220365436 Root MSE = ------------------------------------------------------------------------------ hsgrad | Coef. Std. Err. t P>|t| [95% Conf. Interval]-------------+---------------------------------------------------------------- hsdropout | -1 . . . . . collgrad | -1 . . . . . somecoll | -1 . . . . . _cons | 1 . . . . .------------------------------------------------------------------------------ R 2 =1 so perfect mul:collinearity. How to create a scale . alpha lowincome fstamps tanf femalehd hsdropout unmarried young bllowinc Test scale = mean(unstandardized items) Average interitem covariance: .0320149 Number of items in the scale: 8 Scale reliability coefficient: 0.7086 . egen struggling=rmean(lowincome fstamps tanf femalehd hsdropout unmarried young bllowinc) . sum struggling, detail struggling------------------------------------------------------------- Percentiles Smallest 1% 0 5% 0 10% 0 0 Obs 8690 25% 0 0 Sum of Wgt. 8690 50% .125 Mean .201398 Largest Std. Dev. .2126361 75% .375 1 90% .5 1 Variance .0452141 95% .625 1 Skewness .9276585 99% .75 1 Kurtosis 3.05527 3/16/12 3 Detec:ng Size of Outliers¡ Studen:zed Residuals reg lnfaminc educhd predict r, rstudent sort r list faminc educhd r in 1/10 +-----------------------------+ | faminc educhd r | |-----------------------------| 1. | 0 16 -8.975493 | 2. | 0 16 -8.975493 | 3. | 0 16 -8.975493 | 4. | 0 15 -8.844489 | 5. | 0 15 -8.844489 | |-----------------------------| 6. | 0 14 -8.713711 | 7. | 0 14 -8.713711 | 8. | 0 14 -8.713711 | 9. | 0 13 -8.583151 | 10. | 0 13 -8.583151 | +-----------------------------+ drop if r==....
View Full Document

This note was uploaded on 03/28/2012 for the course PADP 8130 taught by Professor Fertig during the Spring '12 term at LSU.

Page1 / 8

day8prlm - 3/16/12 1 PADP 8130: Linear Models Data Problems...

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

View Full Document Right Arrow Icon
Ask a homework question - tutors are online