------------------------------------------------------------------------------------------------------ log: T:\schmidt-stata\manuf1.log log type: text opened on: 7 Apr 2005, 11:11:10 . regress shipments managers workers materials newcap inventory Source | SS df MS Number of obs = 455 -------------+------------------------------ F( 5, 449) = 5154.49 Model | 9.3059e+16 5 1.8612e+16 Prob > F = 0.0000 Residual | 1.6212e+15 449 3.6108e+12 R-squared = 0.9829 -------------+------------------------------ Adj R-squared = 0.9827 Total | 9.4680e+16 454 2.0855e+14 Root MSE = 1.9e+06 ------------------------------------------------------------------------------ shipments | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- managers | 96.23 6.141296 15.67 0.000 84.16075 108.2993 workers | 16.70028 3.394212 4.92 0.000 10.02976 23.37079 materials | 1.124667 .0148304 75.84 0.000 1.095521 1.153812 newcap | 3.686418 .2701543 13.65 0.000 3.155494 4.217342 inventory | .3656406 .0682858 5.35 0.000 .2314412 .49984 _cons | 258178.2 109967.8 2.35 0.019 42062.62 474293.7 ------------------------------------------------------------------------------ . imtest Cameron & Trivedi's decomposition of IM-test --------------------------------------------------- Source | chi2 df p ---------------------+----------------------------- Heteroskedasticity | 323.15 20 0.0000 Skewness |-3819859.52 5 1.0000 Kurtosis | . 1 . ---------------------+----------------------------- Total | . 26 . --------------------------------------------------- . imtest, white White's test for Ho: homoskedasticity against Ha: unrestricted heteroskedasticity chi2(20) = 323.15 Prob > chi2 = 0.0000 Cameron & Trivedi's decomposition of IM-test --------------------------------------------------- Source | chi2 df p ---------------------+----------------------------- Heteroskedasticity | 323.15 20 0.0000 Skewness |-3819859.52 5 1.0000 Kurtosis | . 1 . ---------------------+----------------------------- Total | . 26 . --------------------------------------------------- . hettest materials Breusch-Pagan / Cook-Weisberg test for heteroskedasticity Ho: Constant variance Variables: materials chi2(1) = 2571.13 Prob > chi2 = 0.0000 . predict res, resid . generate res2=res*res . generate materials2=materials*materials . generate materialssqrt=sqrt(materials) . regress res2 materials2 Source | SS df MS Number of obs = 455 -------------+------------------------------ F( 1, 453) = 91.46 Model | 5.4176e+28 1 5.4176e+28 Prob > F = 0.0000 Residual | 2.6833e+29 453 5.9234e+26 R-squared = 0.1680 -------------+------------------------------ Adj R-squared = 0.1661 Total | 3.2250e+29 454 7.1036e+26 Root MSE = 2.4e+13 ------------------------------------------------------------------------------ res2 | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- materials2 | .0096777 .0010119 9.56 0.000 .007689 .0116664 _cons | 2.50e+12 1.15e+12 2.18 0.030 2.49e+11 4.76e+12 ------------------------------------------------------------------------------ . twoway (scatter res materials) . twoway (scatter res materialssqrt) . twoway (scatter res workers) . twoway (scatter res newcap) . generate shipments1=shipments/materials . generate managers1=managers/materials . generate workers1=workers/materials . generate a1=1/materials . generate newcap1=newcap/materials . generate inventory1=inventory/materials . regress shipments1 managers1 workers1 a1 newcap1 inventory1 Source | SS df MS Number of obs = 455 -------------+------------------------------ F( 5, 449) = 186.78 Model | 143.190256 5 28.6380512 Prob > F = 0.0000 Residual | 68.8414872 449 .153321798 R-squared = 0.6753 -------------+------------------------------ Adj R-squared = 0.6717 Total | 212.031743 454 .467030271 Root MSE = .39156 ------------------------------------------------------------------------------ shipments1 | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- managers1 | 92.22673 6.018158 15.32 0.000 80.39948 104.054 workers1 | 12.47613 2.576051 4.84 0.000 7.413519 17.53875 a1 | -12654.42 7290.849 -1.74 0.083 -26982.85 1673.999 newcap1 | 4.065195 .3698332 10.99 0.000 3.338376 4.792014 inventory1 | .7325114 .1215748 6.03 0.000 .4935852 .9714376 _cons | 1.163453 .0449254 25.90 0.000 1.075163 1.251743 ------------------------------------------------------------------------------ . hettest materials Breusch-Pagan / Cook-Weisberg test for heteroskedasticity Ho: Constant variance Variables: materials chi2(1) = 2.98 Prob > chi2 = 0.0840 . imtest, white White's test for Ho: homoskedasticity against Ha: unrestricted heteroskedasticity chi2(19) = 63.42 Prob > chi2 = 0.0000 Cameron & Trivedi's decomposition of IM-test --------------------------------------------------- Source | chi2 df p ---------------------+----------------------------- Heteroskedasticity | 63.42 19 0.0000 Skewness | 21.42 5 0.0007 Kurtosis | 3.96 1 0.0467 ---------------------+----------------------------- Total | 88.80 25 0.0000 --------------------------------------------------- . generate res1, predict options not allowed r(101); . predict res1, resid . twoway (scatter res1 materials) . twoway (scatter res1 a1) . regress shipments managers workers materials newcap inventory [aweight=materials] (sum of wgt is 1.7507e+09) Source | SS df MS Number of obs = 455 -------------+------------------------------ F( 5, 449) =12612.71 Model | 1.4718e+18 5 2.9436e+17 Prob > F = 0.0000 Residual | 1.0479e+16 449 2.3339e+13 R-squared = 0.9929 -------------+------------------------------ Adj R-squared = 0.9929 Total | 1.4823e+18 454 3.2650e+15 Root MSE = 4.8e+06 ------------------------------------------------------------------------------ shipments | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- managers | 119.025 10.59523 11.23 0.000 98.20261 139.8474 workers | 41.23643 3.773368 10.93 0.000 33.82077 48.65208 materials | 1.216179 .0108623 111.96 0.000 1.194832 1.237527 newcap | 1.038421 .2906638 3.57 0.000 .4671904 1.609651 inventory | -.0673875 .0765979 -0.88 0.379 -.2179223 .0831474 _cons | 4667.451 330572.7 0.01 0.989 -644994.3 654329.2 ------------------------------------------------------------------------------ . regress shipments managers workers materials newcap inventory [aweight=materialssqrt] (sum of wgt is 7.1611e+05) Source | SS df MS Number of obs = 455 -------------+------------------------------ F( 5, 449) = 8513.23 Model | 4.4675e+17 5 8.9350e+16 Prob > F = 0.0000 Residual | 4.7124e+15 449 1.0495e+13 R-squared = 0.9896 -------------+------------------------------ Adj R-squared = 0.9894 Total | 4.5146e+17 454 9.9441e+14 Root MSE = 3.2e+06 ------------------------------------------------------------------------------ shipments | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- managers | 104.1839 7.813839 13.33 0.000 88.82762 119.5401 workers | 27.13721 3.564982 7.61 0.000 20.13109 34.14334 materials | 1.163298 .0121215 95.97 0.000 1.139476 1.18712 newcap | 2.630775 .2805061 9.38 0.000 2.079508 3.182043 inventory | .140807 .0693112 2.03 0.043 .0045923 .2770217 _cons | 275130.2 194558.3 1.41 0.158 -107227.8 657488.1 ------------------------------------------------------------------------------ . regress shipments managers workers materials newcap inventory a1 [aweight=materialssqrt] (sum of wgt is 7.1611e+05) Source | SS df MS Number of obs = 455 -------------+------------------------------ F( 6, 448) = 7083.76 Model | 4.4675e+17 6 7.4459e+16 Prob > F = 0.0000 Residual | 4.7090e+15 448 1.0511e+13 R-squared = 0.9896 -------------+------------------------------ Adj R-squared = 0.9894 Total | 4.5146e+17 454 9.9441e+14 Root MSE = 3.2e+06 ------------------------------------------------------------------------------ shipments | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- managers | 103.8923 7.83638 13.26 0.000 88.49169 119.2929 workers | 26.99094 3.576858 7.55 0.000 19.96144 34.02044 materials | 1.16312 .0121347 95.85 0.000 1.139272 1.186968 newcap | 2.628454 .2807464 9.36 0.000 2.07671 3.180197 inventory | .1389386 .0694406 2.00 0.046 .0024689 .2754083 a1 | -6.97e+10 1.22e+11 -0.57 0.568 -3.10e+11 1.70e+11 _cons | 335755.6 221798.8 1.51 0.131 -100139.6 771650.8 ------------------------------------------------------------------------------ . regress shipments managers workers materials newcap inventory a1 [aweight=materials] (sum of wgt is 1.7507e+09) Source | SS df MS Number of obs = 455 -------------+------------------------------ F( 6, 448) =10487.29 Model | 1.4718e+18 6 2.4530e+17 Prob > F = 0.0000 Residual | 1.0479e+16 448 2.3390e+13 R-squared = 0.9929 -------------+------------------------------ Adj R-squared = 0.9928 Total | 1.4823e+18 454 3.2650e+15 Root MSE = 4.8e+06 ------------------------------------------------------------------------------ shipments | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- managers | 119.0928 10.65336 11.18 0.000 98.15603 140.0295 workers | 41.24918 3.782174 10.91 0.000 33.81617 48.68218 materials | 1.21622 .0108908 111.67 0.000 1.194817 1.237624 newcap | 1.039362 .291313 3.57 0.000 .4668523 1.611872 inventory | -.0671319 .0767743 -0.87 0.382 -.2180143 .0837505 a1 | 3.05e+10 4.47e+11 0.07 0.946 -8.48e+11 9.09e+11 _cons | -9450.205 390245.9 -0.02 0.981 -776390 757489.6 ------------------------------------------------------------------------------ . regress shipments managers workers materials newcap inventory a1 [aweight=materials2] (sum of wgt is 4.9878e+16) Source | SS df MS Number of obs = 455 -------------+------------------------------ F( 6, 448) =14929.74 Model | 2.1428e+18 6 3.5714e+17 Prob > F = 0.0000 Residual | 1.0717e+16 448 2.3921e+13 R-squared = 0.9950 -------------+------------------------------ Adj R-squared = 0.9950 Total | 2.1535e+18 454 4.7435e+15 Root MSE = 4.9e+06 ------------------------------------------------------------------------------ shipments | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- managers | 174.4245 17.76876 9.82 0.000 139.5041 209.345 workers | 59.43217 4.124432 14.41 0.000 51.32653 67.53781 materials | 1.307009 .0088762 147.25 0.000 1.289565 1.324453 newcap | -1.405297 .2672414 -5.26 0.000 -1.930499 -.8800946 inventory | -.3559808 .0963619 -3.69 0.000 -.5453583 -.1666033 a1 | 6.28e+12 3.17e+12 1.98 0.048 5.40e+10 1.25e+13 _cons | -3041124 858591.8 -3.54 0.000 -4728492 -1353757 ------------------------------------------------------------------------------ . use "T:\schmidt-stata\schoolspend.dta", clear . regress totalspend totalstateaid income minority Source | SS df MS Number of obs = 50 -------------+------------------------------ F( 3, 46) = 213.91 Model | 1.4757e+09 3 491908574 Prob > F = 0.0000 Residual | 105780023 46 2299565.72 R-squared = 0.9331 -------------+------------------------------ Adj R-squared = 0.9288 Total | 1.5815e+09 49 32275627.4 Root MSE = 1516.4 ------------------------------------------------------------------------------ totalspend | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- totalstate~d | -.2778458 .3049163 -0.91 0.367 -.8916104 .3359188 income | .0320546 .0045188 7.09 0.000 .0229587 .0411505 minority | -7.01137 19.05167 -0.37 0.715 -45.36038 31.33764 _cons | 207.5894 392.4485 0.53 0.599 -582.3685 997.5472 ------------------------------------------------------------------------------ . hettest population Breusch-Pagan / Cook-Weisberg test for heteroskedasticity Ho: Constant variance Variables: population chi2(1) = 140.36 Prob > chi2 = 0.0000 . imtest, white White's test for Ho: homoskedasticity against Ha: unrestricted heteroskedasticity chi2(9) = 25.42 Prob > chi2 = 0.0025 Cameron & Trivedi's decomposition of IM-test --------------------------------------------------- Source | chi2 df p ---------------------+----------------------------- Heteroskedasticity | 25.42 9 0.0025 Skewness | 4.72 3 0.1934 Kurtosis | 1.29 1 0.2568 ---------------------+----------------------------- Total | 31.43 13 0.0029 --------------------------------------------------- . predict r, resid . twoway (scatter r population) . - preserve . twoway (scatter r totalstateaid) . log close log: T:\schmidt-stata\manuf1.log log type: text closed on: 7 Apr 2005, 11:48:50 ------------------------------------------------------------------------------------------------------