|
|
Because of the nature of the tobit model,
a standard R2 isn't calculated. However, you can calculate a pseudo-R2
by calculating pseudo residuals and using them in the usual R2 formula.
The tricky part is calculating the pseudo predicted values to get
the residuals. It's hard to write the formula in plain email, so
you might want to Google for it, but here's a try.
yhat = X*beta * N(X*beta/sigma) - sigma
* n(X*beta/sigma)
where N is the Normal CDF, n the normal
density, beta and sigma the estimates, and the observation subscript is
omitted.
HTH,
Jon Peck
Senior Software Engineer, IBM
[hidden email]
312-651-3435
From:
Inga <[hidden email]>
To:
[hidden email]
Date:
10/27/2010 06:50 AM
Subject:
[SPSSX-L] R2
calculation in Tobit regression
Sent by:
"SPSSX(r)
Discussion" <[hidden email]>
Hello everybody,
I was able to install the R Extension Command for Tobit Regression ole
ole.
Everything works fine so far.
However, i am wondering now how to calculate R-squared. Can anyone help
me
please?
Best
Inga
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/R2-calculation-in-Tobit-regression-tp3238416p3238416.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
|
Locked
