Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

[no subject]


From   Ehsan Rahman <ehrahman@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Date   Wed, 14 Nov 2007 13:24:43 -0800 (PST)

Hi I am trying to do Terza's 2-step method of moments estimation in Stata, however I
am getting a strange error message. When I run it as it is below, with "1-normal(t)"
in the denominator of the second term in the correction equation, I get the following
error: "starting values invalid or some RHS variables have missing values." 

If however, I simlply change "1-normal(t)" to ".99999-normal(t)" the estimation runs
as expected. If anyone can explain the cause of this error, I would be very greatful.
Below are the two versions of the estimation as well as links to the relevant files.
Thank you for your help.

- Ehsan Rahman

//This works properly

capture drop t

probit PRIVINS RGN2 RGN3 RGN4 MSA AGE AGE2 SEX WHITE MAR UNIV INC EMP SICKPAY ILLNESS
DISABILITY DISEASE

gen t = _b[_cons] + _b[RGN2]*RGN2 +_b[RGN3]*RGN3 +_b[RGN4]*RGN4 +_b[MSA]*MSA
+_b[AGE]*AGE +_b[AGE2]*AGE2 +_b[SEX]*SEX +_b[WHITE]*WHITE +_b[MAR]*MAR +_b[UNIV]*UNIV
+_b[INC]*INC +_b[EMP]*EMP +_b[SICKPAY]*SICKPAY +_b[ILLNESS]*ILLNESS
+_b[DISABILITY]*DISABILITY +_b[DISEASE]*DISEASE

nl
(PHY=exp({B0}+{B1}*RGN2+{B2}*RGN3+{B3}*RGN4+{B4}*MSA+{B5}*AGE+{B6}*AGE2+{B7}*SEX+{B8}*WHITE+{B9}*MAR+{B10}*UNIV+{B11}*INC+{B12}*EMP+{B13}*SICKPAY+{B14}*ILLNESS+{B15}*DISABILITY+{B16}*DISEASE+{B17}*PUBINS+{B18}*PRIVINS)*(
 (PRIVINS*normal({B19} + t)/normal(t)  ) +
((1-PRIVINS)*(1-normal({B19}+t))/(.99999-normal(t))   )     )          )

//This version produces the error message

capture drop t

probit PRIVINS RGN2 RGN3 RGN4 MSA AGE AGE2 SEX WHITE MAR UNIV INC EMP SICKPAY ILLNESS
DISABILITY DISEASE

gen t = _b[_cons] + _b[RGN2]*RGN2 +_b[RGN3]*RGN3 +_b[RGN4]*RGN4 +_b[MSA]*MSA
+_b[AGE]*AGE +_b[AGE2]*AGE2 +_b[SEX]*SEX +_b[WHITE]*WHITE +_b[MAR]*MAR +_b[UNIV]*UNIV
+_b[INC]*INC +_b[EMP]*EMP +_b[SICKPAY]*SICKPAY +_b[ILLNESS]*ILLNESS
+_b[DISABILITY]*DISABILITY +_b[DISEASE]*DISEASE

nl
(PHY=exp({B0}+{B1}*RGN2+{B2}*RGN3+{B3}*RGN4+{B4}*MSA+{B5}*AGE+{B6}*AGE2+{B7}*SEX+{B8}*WHITE+{B9}*MAR+{B10}*UNIV+{B11}*INC+{B12}*EMP+{B13}*SICKPAY+{B14}*ILLNESS+{B15}*DISABILITY+{B16}*DISEASE+{B17}*PUBINS+{B18}*PRIVINS)*(
 (PRIVINS*normal({B19} + t)/normal(t)  ) +
((1-PRIVINS)*(1-normal({B19}+t))/(1-normal(t))   )     )          )


//relevant files

https://mavspace.uta.edu:443/exr5618/ins2.dta
https://mavspace.uta.edu:443/exr5618/terza5.do
https://mavspace.uta.edu:443/exr5618/log2.txt




      ____________________________________________________________________________________
Be a better pen pal. 
Text or chat with friends inside Yahoo! Mail. See how.  http://overview.mail.yahoo.com/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index