------------------------------------------------------------------------------------- log: N:\www\econ30331\productivity.log log type: text opened on: 17 Mar 2010, 09:16:18 . . . tsset time time variable: time, 1 to 61 delta: 1 unit . . * generate ln of outcomes . gen rwagel=ln(rwage) . gen prodl=ln(productivity) . . * run a model of productivity . * on just a time trend . * see how well it fits . reg prodl time Source | SS df MS Number of obs = 61 -------------+------------------------------ F( 1, 59) = 4128.39 Model | 7.79439507 1 7.79439507 Prob > F = 0.0000 Residual | .1113919 59 .001887998 R-squared = 0.9859 -------------+------------------------------ Adj R-squared = 0.9857 Total | 7.90578697 60 .131763116 Root MSE = .04345 ------------------------------------------------------------------------------ prodl | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- time | .0203023 .000316 64.25 0.000 .01967 .0209346 _cons | 3.684954 .0112649 327.12 0.000 3.662413 3.707495 ------------------------------------------------------------------------------ . * output predicted value . predict prodl_pred (option xb assumed; fitted values) . . *output predicted values to csv . * file to graph in pretty graph . outsheet time year prodl prodl_pred using predict_value.csv, comma . . . . reg rwagel time prodl Source | SS df MS Number of obs = 61 -------------+------------------------------ F( 2, 58) = 2083.25 Model | 4.65023272 2 2.32511636 Prob > F = 0.0000 Residual | .06473372 58 .001116099 R-squared = 0.9863 -------------+------------------------------ Adj R-squared = 0.9858 Total | 4.71496644 60 .078582774 Root MSE = .03341 ------------------------------------------------------------------------------ rwagel | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- time | -.0147661 .0020467 -7.21 0.000 -.018863 -.0106692 prodl | 1.479204 .1000978 14.78 0.000 1.278837 1.679572 _cons | -1.531055 .3689574 -4.15 0.000 -2.269604 -.7925068 ------------------------------------------------------------------------------ . estat dwatson Durbin-Watson d-statistic( 3, 61) = .3270161 . . . * test for random walk . * first, generate lags . . gen rwagel1=rwagel[_n-1] (1 missing value generated) . gen prodl1=prodl[_n-1] (1 missing value generated) . . . * regress values on lags . * test that coef on lag equals1 . reg rwagel time rwagel1 Source | SS df MS Number of obs = 60 -------------+------------------------------ F( 2, 57) =12947.36 Model | 4.32027231 2 2.16013616 Prob > F = 0.0000 Residual | .009509872 57 .00016684 R-squared = 0.9978 -------------+------------------------------ Adj R-squared = 0.9977 Total | 4.32978218 59 .073386139 Root MSE = .01292 ------------------------------------------------------------------------------ rwagel | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- time | .0004911 .0003717 1.32 0.192 -.0002532 .0012354 rwagel1 | .9446265 .0233767 40.41 0.000 .8978154 .9914375 _cons | .2443952 .0912928 2.68 0.010 .0615845 .4272059 ------------------------------------------------------------------------------ . test rwagel1=1 ( 1) rwagel1 = 1 F( 1, 57) = 5.61 Prob > F = 0.0213 . . . reg prodl time prodl1 Source | SS df MS Number of obs = 60 -------------+------------------------------ F( 2, 57) =18572.67 Model | 7.39083745 2 3.69541873 Prob > F = 0.0000 Residual | .011341335 57 .000198971 R-squared = 0.9985 -------------+------------------------------ Adj R-squared = 0.9984 Total | 7.40217879 59 .125460657 Root MSE = .01411 ------------------------------------------------------------------------------ prodl | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- time | .001841 .0008647 2.13 0.038 .0001095 .0035726 prodl1 | .9011966 .0422653 21.32 0.000 .8165617 .9858315 _cons | .389113 .1549295 2.51 0.015 .0788721 .699354 ------------------------------------------------------------------------------ . test prodl1=1 ( 1) prodl1 = 1 F( 1, 57) = 5.46 Prob > F = 0.0229 . . * now difference data . . gen drwagel=rwagel-rwagel1 (1 missing value generated) . gen dprodl=prodl-prodl1 (1 missing value generated) . . * run regression of diff on diff . reg drwagel time dprodl Source | SS df MS Number of obs = 60 -------------+------------------------------ F( 2, 57) = 17.71 Model | .004894165 2 .002447082 Prob > F = 0.0000 Residual | .007874197 57 .000138144 R-squared = 0.3833 -------------+------------------------------ Adj R-squared = 0.3617 Total | .012768361 59 .000216413 Root MSE = .01175 ------------------------------------------------------------------------------ drwagel | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- time | -.000284 .0000893 -3.18 0.002 -.0004629 -.0001051 dprodl | .4548905 .1054274 4.31 0.000 .2437759 .6660052 _cons | .0159993 .0042484 3.77 0.000 .0074921 .0245065 ------------------------------------------------------------------------------ . estat dwatson Durbin-Watson d-statistic( 3, 60) = 1.5944 . prais drwagel time dprodl, corc twostep Iteration 0: rho = 0.0000 Iteration 1: rho = 0.1679 Cochrane-Orcutt AR(1) regression -- twostep estimates Source | SS df MS Number of obs = 59 -------------+------------------------------ F( 2, 56) = 16.81 Model | .004261964 2 .002130982 Prob > F = 0.0000 Residual | .007100571 56 .000126796 R-squared = 0.3751 -------------+------------------------------ Adj R-squared = 0.3528 Total | .011362534 58 .000195906 Root MSE = .01126 ------------------------------------------------------------------------------ drwagel | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- time | -.0003332 .0001048 -3.18 0.002 -.000543 -.0001233 dprodl | .4298405 .1003206 4.28 0.000 .2288741 .6308069 _cons | .0186114 .0046416 4.01 0.000 .0093132 .0279096 -------------+---------------------------------------------------------------- rho | .1679271 ------------------------------------------------------------------------------ Durbin-Watson statistic (original) 1.594400 Durbin-Watson statistic (transformed) 1.827364 . . . end of do-file . log close log: N:\www\econ30331\productivity.log log type: text closed on: 17 Mar 2010, 09:16:25 -------------------------------------------------------------------------------------