Hello,
Would anyone be able to help me with the following question--I would really appreciate it!
I am analyzing a choice between 3 types of mortgage products, and the iir is consistenlty rejected by all tests. I have seen several sources saying that multinomial probit can be used instead of multinomial logit in this case-- but all the examples of multinomial probit I see have to do with joint decisions (such as voting for taxes and using public schools) rather than chosing between, in this case, different types of mortgage products.
As a practical matter, multinomial probit gives estimates similar to multinomial logit...
is it appropriate in this case?
Thanks so much,
Irina
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Joao Ricardo F. Lima
Sent: Thursday, April 09, 2009 5:37 PM
To: [email protected]
Subject: Re: st: RE: question from statalist
Irina,
try:
- margeff, at(mean) replace
My version of margeff works fine:
. which margeff
c:\ado\plus\m\margeff.ado
*! Obtain partial effects after estimation
*! Version 2.1.8 (20 April 2008) (Revision of Stata Journal submission)
*! Author: Tamas Bartus (Corvinus University, Budapest)
HTH,
Joao Lima
2009/4/9 Paley, Irina <[email protected]>:
> Just to see if it runs, I dropped all states with less than 2000 observations. Now all the SE are defined...and it doesn't look like stata drops any additional states because of multicollinearity... But it gives the same error:
>
> . xi3: mlogit prod_type female_o ///
>> fico ltv dti income loanamount loanterm /// ba_new f_min_arm5
>> t10_min_t1 e.state
> e.state _Istate_4-53 (naturally coded; _Istate_4
> omitted)
>
> Iteration 0: log likelihood = -72881.502 Iteration 1: log
> likelihood = -55904.913 Iteration 2: log likelihood = -49007.083
> Iteration 3: log likelihood = -47984.956 Iteration 4: log
> likelihood = -47889.198 Iteration 5: log likelihood = -47887.349
> Iteration 6: log likelihood = -47887.345 Iteration 7: log
> likelihood = -47887.345
>
> Multinomial logistic regression Number of obs =
> 133586
> LR chi2(60) =
> 49988.31
> Prob > chi2 =
> 0.0000 Log likelihood = -47887.345 Pseudo R2
> = 0.3429
>
> ----------------------------------------------------------------------
> --------
> prod_type | Coef. Std. Err. z P>|z| [95% Conf.
> Interval]
> -------------+--------------------------------------------------------
> -------------+--------
> p2 |
> female_only | -.1555133 .0297789 -5.22 0.000 -.2138789
> -.0971477
> fico | -.000158 .0002554 -0.62 0.536 -.0006586
> .0003427
> ltv | -.0117513 .0010367 -11.34 0.000 -.0137832
> -.0097195
> dti | .0010724 .0012078 0.89 0.375 -.0012948
> .0034395
> income | .0001198 .0000953 1.26 0.209 -.000067
> .0003067
> loanamount | .0027244 .0000848 32.13 0.000 .0025581
> .0028906
> loanterm | .3048393 .0059386 51.33 0.000 .2931999
> .3164787
> ba_new | 1.380727 .1570039 8.79 0.000 1.073006
> 1.688449
> f_min_arm5 | 3.34822 .2489538 13.45 0.000 2.860279
> 3.836161
> t10_min_t1 | -.2526719 .0454174 -5.56 0.000 -.3416883
> -.1636555
> _Istate_6 | 1.177903 .0375687 31.35 0.000 1.104269
> 1.251536
> _Istate_9 | -.4211162 .1363362 -3.09 0.002 -.6883301
> -.1539022
> _Istate_12 | .1748989 .0508363 3.44 0.001 .0752615
> .2745363
> _Istate_13 | .0120183 .0621117 0.19 0.847 -.1097183
> .1337549
> _Istate_17 | .5266333 .0624313 8.44 0.000 .4042702
> .6489965
> _Istate_20 | -.6538387 .1589123 -4.11 0.000 -.9653011
> -.3423762
> _Istate_24 | .7937781 .0586172 13.54 0.000 .6788905
> .9086657
> _Istate_25 | -.2993312 .0892851 -3.35 0.001 -.4743268
> -.1243355
> _Istate_29 | -.6722612 .1141442 -5.89 0.000 -.8959798
> -.4485427
> _Istate_32 | .9195407 .0929692 9.89 0.000 .7373244
> 1.101757
> _Istate_34 | -.5387866 .1017824 -5.29 0.000 -.7382764
> -.3392967
> _Istate_36 | -.6020304 .0828407 -7.27 0.000 -.7643951
> -.4396657
> _Istate_37 | -.2893397 .0708244 -4.09 0.000 -.428153
> -.1505265
> _Istate_40 | -.9929478 .1939296 -5.12 0.000 -1.373043
> -.6128529
> _Istate_42 | -.4063065 .1287868 -3.15 0.002 -.658724
> -.1538889
> _Istate_45 | .0057848 .082172 0.07 0.944 -.1552694
> .1668389
> _Istate_47 | -.5059151 .1443456 -3.50 0.000 -.7888274
> -.2230028
> _Istate_48 | -.7562123 .0631341 -11.98 0.000 -.8799529
> -.6324718
> _Istate_51 | .6622584 .0612535 10.81 0.000 .5422037
> .782313
> _Istate_53 | .8346187 .0676625 12.34 0.000 .7020027
> .9672348
> _cons | -13.33689 .3214546 -41.49 0.000 -13.96693
> -12.70686
> -------------+--------------------------------------------------------
> -------------+--------
> p3 |
> female_only | .2520734 .0216934 11.62 0.000 .2095551
> .2945916
> fico | -.005479 .0001709 -32.06 0.000 -.0058139
> -.0051441
> ltv | .1089525 .0014794 73.65 0.000 .106053
> .111852
> dti | -.0131825 .0011063 -11.92 0.000 -.0153509
> -.0110141
> income | -.0097414 .0003657 -26.64 0.000 -.0104581
> -.0090246
> loanamount | .0053394 .0001353 39.46 0.000 .0050742
> .0056046
> loanterm | -.30474 .0031175 -97.75 0.000 -.3108501
> -.2986298
> ba_new | 1.0459 .1243753 8.41 0.000 .8021284
> 1.289671
> f_min_arm5 | 1.399249 .1952819 7.17 0.000 1.016503
> 1.781994
> t10_min_t1 | -1.661645 .0393439 -42.23 0.000 -1.738758
> -1.584533
> _Istate_6 | 2.378287 .032999 72.07 0.000 2.31361
> 2.442964
> _Istate_9 | -.1839164 .1083993 -1.70 0.090 -.3963751
> .0285423
> _Istate_12 | 1.200839 .0346214 34.68 0.000 1.132983
> 1.268696
> _Istate_13 | -.0277285 .0458872 -0.60 0.546 -.1176658
> .0622087
> _Istate_17 | .0207768 .0673345 0.31 0.758 -.1111964
> .1527499
> _Istate_20 | -2.170635 .1792299 -12.11 0.000 -2.521919
> -1.819351
> _Istate_24 | 1.474792 .043018 34.28 0.000 1.390479
> 1.559106
> _Istate_25 | .1024874 .0684801 1.50 0.134 -.0317311
> .2367059
> _Istate_29 | -.6156074 .0730468 -8.43 0.000 -.7587765
> -.4724384
> _Istate_32 | 2.030276 .0636586 31.89 0.000 1.905507
> 2.155044
> _Istate_34 | .1079523 .0672633 1.60 0.109 -.0238814
> .239786
> _Istate_36 | -.0162838 .0647249 -0.25 0.801 -.1431423
> .1105746
> _Istate_37 | -.6099104 .0532916 -11.44 0.000 -.7143601
> -.5054607
> _Istate_40 | -3.112063 .2462253 -12.64 0.000 -3.594656
> -2.62947
> _Istate_42 | -1.117306 .1131833 -9.87 0.000 -1.339141
> -.895471
> _Istate_45 | -.6461894 .0765761 -8.44 0.000 -.7962758
> -.4961031
> _Istate_47 | -1.349594 .1318826 -10.23 0.000 -1.608079
> -1.091109
> _Istate_48 | -2.136153 .0655895 -32.57 0.000 -2.264706
> -2.0076
> _Istate_51 | 1.383357 .044693 30.95 0.000 1.295761
> 1.470954
> _Istate_53 | 1.305855 .0550159 23.74 0.000 1.198026
> 1.413684
> _cons | -.8105805 .213756 -3.79 0.000 -1.229535
> -.3916265
> ----------------------------------------------------------------------
> --------
> (prod_type==p1 is the base outcome)
>
> margeff
> invalid syntax
>
>
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Maarten
> buis
> Sent: Thursday, April 09, 2009 12:05 PM
> To: [email protected]
> Subject: Re: st: RE: question from statalist
>
>
> --- On Wed, 8/4/09, Paley, Irina wrote:
>> I use your exact code for mlogit, adapted to my dataset, and when I
>> run margeff I get that it's invalid syntax. What do you mean by
>> leaving out reference category?
>
> Say you want to controll for gender, than there are two dummies:
> male (indicating who is male) and female (indicating who is female). These two variables contain superfluous information (if you know that someone isn't male than she is probably a female).
> Because of that you can't add both variables in a regression, and you need to leave one of these two out of your model. The variable you leave out is called the reference category. The same is true for your state dummies, you need to leave one of the states out.
>
> As you created your dummies using -xi3- this already happened, so this is not the problem. However, given that some of the state dummies have missing values on the standard errors suggests that there is still a problem with your model: you just don't have enough information in your data to add all the state dummies.
> The easiest solution would be if you could find nearby states that are
> sort of similar to the problematic states, and merge these into "super
> states". So, States 50 54 2 30 46 and 72 are problematic (they either
> have a missing value on the standard error or they have been dropped
> outright due to
> multiconlinearity) and asssume that state 50 is close to State 51, and state 54 to state 3 . Than you can create the "supper states" by making a copy of the variable state and use -recode-:
>
> gen state2 = state
> recode state2 (50=51) (54=3) etc.
>
> And than you use state2 instead of state in your -mlogit- model.
>
> Hope this helps,
> Maarten
>
> -----------------------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
> http://home.fsw.vu.nl/m.buis/
> -----------------------------------------
>
>
>
>
> *
> * 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/
>
> *
> * 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/
>
--
----------------------------------------
Joao Ricardo Lima, D.Sc.
Professor
UFPB-CCA-DCFS
Fone: +5538387264913
Skype: joao_ricardo_lima
----------------------------------------
*
* 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/
*
* 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/
