In my regression, the dependent variable is log-transformed (wdealsize).
Code:
Linear regression Number of obs = 150 F(35, 73) = 14.50 Prob > F = 0.0000 R-squared = 0.3877 Root MSE = 2.0892 (Std. Err. adjusted for 74 clusters in companyno) --------------------------------------------------------------------------------- | Robust wdealsize | Coef. Std. Err. t P>|t| [95% Conf. Interval] ----------------+---------------------------------------------------------------- lagldf | 4.528605 2.370172 1.91 0.060 -.1951407 9.252351 lagprof | 13.36013 3.440691 3.88 0.000 6.502839 20.21741 lagmtb | 1.350247 .5355359 2.52 0.014 .282925 2.417568 lagsize | 1.157787 .4421126 2.62 0.011 .2766577 2.038916 lagsr | .1761364 .1127095 1.56 0.122 -.0484934 .4007662 lagcr | 1.1887 1.86192 0.64 0.525 -2.522101 4.899501 1.serialac | -.5547087 .3020776 -1.84 0.070 -1.156749 .0473311 1.crossbordern | .0602353 .4559033 0.13 0.895 -.8483787 .9688492 1.targetpublicn | .6101345 .4263404 1.43 0.157 -.2395607 1.45983 1.diversn | .2126942 .4710001 0.45 0.653 -.7260076 1.151396
Code:
. display exp(4.528605) 92.629253
Many thanks.
0 Response to Large coefficients in a regression where the dependent variable is log-transformed
Post a Comment