Hi,
I am new to STATA and am currently doing (trying) multiple regression using dummy coded categorical variables, I want to compare my coefficients so I can assess the relative strength of each predictor as I would in multiple regression with continuous predictors. is it possible to get an overall coefficient for my dummy coded variables?
. regress Group27_weight highested_22 highested_23 highested_24 BMI_M2 BMI_M3, beta
Source | SS df MS Number of obs = 1602
-------------+------------------------------ F( 5, 1596) = 8.74
Model | 512485.705 5 102497.141 Prob> F = 0.0000
Residual | 18722950.7 1596 11731.1721 R-squared = 0.0266
-------------+------------------------------ Adj R-squared = 0.0236
Total | 19235436.4 1601 12014.6386 Root MSE = 108.31
------------------------------------------------------------------------------
Group27_we~t | Coef. Std. Err. t P>|t| Beta
-------------+----------------------------------------------------------------
highested_22 | 24.77708 11.51337 2.15 0.032 .1130292
highested_23 | 23.56448 12.14225 1.94 0.052 .0935415
highested_24 | 62.28482 12.72192 4.90 0.000 .2149551
BMI_M2 | -9.841873 6.368431 -1.55 0.122 -.0395633
BMI_M3 | -16.02685 7.922844 -2.02 0.043 -.0518379
_cons | 149.9236 11.31615 13.25 0.000 .
------------------------------------------------------------------------------
e.g. is it possible to assess if highested_2 or BMI_M has the largest coefficient? or standardised coefficent?
Many thanks in advance,
kind regards
_________________________________________________________________
Save time by using Hotmail to access your other email accounts.
http://clk.atdmt.com/UKM/go/167688463/direct/01/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
