According to its help the .ado-program -fitstat- (source see below) should calculate McKelvey-Zavoina's R² for binary logit models. This worked fine until one of my latest updates of .ado-programs. Now, however, it will only calculate McKelvey-Zavoina's R² as a post-estimation command after -logit-, not anymore after -logistic-:
Is there anybody with a version of -fitstat- that will calculate McKelvey-Zavoina's R² after -logistic-? Which version is it and from where did you install it? Or is there anybody who can suggest a modification to the the current version (see below) of fitstat.ado to produce this statistic after -logistic-, as well? I had a look at the source code, but couldn't find the correct place for a respective modification.
The .ado-program -fitstat- is part of several different packages (and its versions) and it is not easy to find or update to the most recent one. Here I am referring to -fitstat- that is part of the package spost13_ado as it is available from "https://jslsoc.sitehost.iu.edu/stata" via
with version number
Code:
. sysuse auto, clear
(1978 Automobile Data)
. logit foreign price
Iteration 0: log likelihood = -45.03321
Iteration 1: log likelihood = -44.947363
Iteration 2: log likelihood = -44.94724
Iteration 3: log likelihood = -44.94724
Logistic regression Number of obs = 74
LR chi2(1) = 0.17
Prob > chi2 = 0.6784
Log likelihood = -44.94724 Pseudo R2 = 0.0019
------------------------------------------------------------------------------
foreign | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
price | .0000353 .0000844 0.42 0.676 -.0001301 .0002006
_cons | -1.079792 .5878344 -1.84 0.066 -2.231927 .0723419
------------------------------------------------------------------------------
. fitstat
| logit
-------------------------+-------------
Log-likelihood |
Model | -44.947
Intercept-only | -45.033
-------------------------+-------------
Chi-square |
Deviance(df=72) | 89.894
LR(df=1) | 0.172
p-value | 0.678
-------------------------+-------------
R2 |
McFadden | 0.002
McFadden(adjusted) | -0.043
McKelvey & Zavoina | 0.003
Cox-Snell/ML | 0.002
Cragg-Uhler/Nagelkerke | 0.003
Efron | 0.002
Tjur's D | 0.002
Count | 0.703
Count(adjusted) | 0.000
-------------------------+-------------
IC |
AIC | 93.894
AIC divided by N | 1.269
BIC(df=2) | 98.503
-------------------------+-------------
Variance of |
e | 3.290
y-star | 3.301
. logistic foreign price
Logistic regression Number of obs = 74
LR chi2(1) = 0.17
Prob > chi2 = 0.6784
Log likelihood = -44.94724 Pseudo R2 = 0.0019
------------------------------------------------------------------------------
foreign | Odds Ratio Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
price | 1.000035 .0000844 0.42 0.676 .9998699 1.000201
_cons | .339666 .1996674 -1.84 0.066 .1073214 1.075023
------------------------------------------------------------------------------
Note: _cons estimates baseline odds.
. fitstat
| logistic
-------------------------+-------------
Log-likelihood |
Model | -44.947
Intercept-only | -45.033
-------------------------+-------------
Chi-square |
Deviance(df=72) | 89.894
LR(df=1) | 0.172
p-value | 0.678
-------------------------+-------------
R2 |
McFadden | 0.002
McFadden(adjusted) | -0.043
Cox-Snell/ML | 0.002
Cragg-Uhler/Nagelkerke | 0.003
Efron | 0.002
Tjur's D | 0.002
Count | 0.703
Count(adjusted) | 0.000
-------------------------+-------------
IC |
AIC | 93.894
AIC divided by N | 1.269
BIC(df=2) | 98.503
The .ado-program -fitstat- is part of several different packages (and its versions) and it is not easy to find or update to the most recent one. Here I am referring to -fitstat- that is part of the package spost13_ado as it is available from "https://jslsoc.sitehost.iu.edu/stata" via
Code:
net install spost13_ado, from(https://jslsoc.sitehost.iu.edu/stata)
Code:
*! 4.1.10 | 2017-01-05 | freese long | fixed bug in tobit (for v <14 and 15+)
Comment