Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# st: Krinsky Robb method for WTP estimates from existing varcov matrix and estimates

 From "sanja_lutzeyer@ncsu.edu" To statalist@hsphsun2.harvard.edu Subject st: Krinsky Robb method for WTP estimates from existing varcov matrix and estimates Date Tue, 13 Nov 2012 09:58:04 -0500

```Good day.
I am currently using a program called Latent Gold to do latent class
analysis on my data.
My main interest, however, is the welfare analysis associated with the
coefficient estimates, not the estimates themselves. Latent Gold does
not in itself allow for post-estimation analysis (as far as I know.)
I would like to use the Krinsky and Robb method to estimate standard
errors and confidence intervals.
I have figured out how to do this when running a simply logit model in
STATA and saving the results afterwards, but I can't seem to figure
out how to do it in STATA
I have managed to export the variance covariance matrix in one file
into STATA. The variables are named, and below them is a 50x50 matrix
of numbers.
I then have a separate STATA file in which I have the same variable
names and underneath them the coefficient estiamtes of these 50
variables (i.e. a 1x50 vector).
I have managed to append the data files so that the last row under the
variable names (the 51st row) represents the coefficient.

I then did the following:
mkmat (I put the 50 variables here), matrix(total)
matrix varcov = total[1...50,1...]
matrix b = total[51,1...]

To use the Krinsky-Robb method, I then tried to draw norms for the variables...

drawnorm  x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 x13 x14 x15 x16 x17
x18 x19 x20 x21 x22 x23 x24 x25 x26 x27 x28 x29 x30 x31 x32 x33 x34
x35 x36 x37 x38 x39 x40 x41 x42 x43 x44 x45 x46 x47 x48 x49 x50,
n(10000) means(b) cov(varcov) cstorage(full) clear

but then I get the warning
option cstorage() invalid; vectorized corr/cov matrix found. I'm not
sure what I'm doing wrong.

Once this step was done, I was planning on simply generating the
willingness to pay measures for the variables by, for example:
generate wtp5 = (x1/-x43) as -x43 represents the price variable.

Any suggestions would be much appreciated.
If any suggestions for the Delta method exist, that would also be
great. I am only using the Krinsky-Robb as I am not sure how to code
the Delta method in STATA.