Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# st: Varying results for levpet

 From SAM MCCAW To statalist@hsphsun2.harvard.edu Subject st: Varying results for levpet Date Fri, 5 Oct 2012 17:38:52 -0400

```Hi All,

I have run into an issue with my stata results when I run levpet
industry by industry.

When I run the following command for Levihnson-Petrin I get different
results each time.

eststo: levpet lnY if industry_no == 1, free (lnl) proxy (lnM) capital
(lnK1) revenue

For example, the first and second set of results are posted below. Is
this a similar effect to bootstrapping?

Thanks a lot.

SAM

//1.

Levinsohn-Petrin productivity estimator

Dependent variable represents revenue.          Number of obs      =      2042
Group variable (i): id                                      Number of
groups   =     58591
Time variable (t): year
Obs
per group: min =         1

avg =       1.2

max =         3

------------------------------------------------------------------------------
lnY |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
lnl |     .1448912   .0225324     6.43   0.000     .1007285    .1890538
lnK1 |   .0336222   .0698806     0.48   0.630    -.1033412    .1705856
lnM |               1   .2034656     4.91   0.000
.6012147    1.398785
------------------------------------------------------------------------------
Wald test of constant returns to scale: Chi2 =   0.97 (p = 0.3243).
(est1 stored)

.
//2.

Levinsohn-Petrin productivity estimator

Dependent variable represents revenue.          Number of obs      =      2042
Group variable (i): id                          Number of groups   =     58591
Time variable (t): year
Obs per group: min =         1
avg =       1.2
max =         3

------------------------------------------------------------------------------
lnY |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
lnl |     .1448912   .0273666     5.29   0.000     .0912537    .1985287
lnK1 |   .0336222    .066174     0.51    0.611    -.0960764    .1633209
lnM |              1   .2253513     4.44    0.000
.5583196     1.44168
------------------------------------------------------------------------------
Wald test of constant returns to scale: Chi2 =   0.95 (p = 0.3299).
(est2 stored)

.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/
```