Statalist


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

Re: st: -est combine-?


From   Michael Hanson <mshanson@mac.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: -est combine-?
Date   Mon, 30 Jul 2007 15:50:15 -0400

On Jul 30, 2007, at 10:30 AM, Maarten buis wrote:
Oops, -estadd- is part of -estout-, so see -ssc desc estout- and look
for -estadd-.

--- Maarten Buis <M.Buis@fsw.vu.nl> wrote:

-ssc desc estadd-
Thanks, Maarten. As I noted previously (and detail below), I had attempted to use the tools in the -estout- suite without success. I would, of course, welcome more specific suggestions on the application of -estadd- to my problem.


On Jul 30, 2007, at 10:30 AM, Ben Jann wrote:


Michael,
have a look at

 http://www.stata.com/statalist/archive/2007-06/msg00636.html

Not exactly your problem, but maybe it helps.
Thanks, Ben. I'm looking at that now, trying to see if I can re- purpose it for my application.



Furthermore, the -erepost- command might be helpful (see -ssc describe erepost-). The
procedure would be to do something as follows:

ivreg2 ...
est sto base
matrix b = e(b)
matrix V = e(V)
nlcom ...
... add results to b and V
[... more nlcom commands ...]
est restore base
erepost b=b V=V
est sto modified
estout modified
Thanks again; I'm also looking into whether I can get -erepost- to do my bidding. Both of these approaches are just beyond my current understanding of how to manipulate -estimates- results, but I plan to take a stab at them today.



If you provide more details, I can be more specific.
ben
Here is a set of examples that may clarify my question. Given the notation below, I'd ultimately like one table with /a0, /b0, /b1, / b2, /d1, and /d2 (see the end of the example) listed in a single column, along with the p-values (or HAC SEs) for each of these six estimated parameters. My apologies in advance for the length of this message.


// Regression

. nl (y = (1-{a0})*({b0} + {b1}*x1 + {b2}*x2) + {a0}*Ly), vce(hac nw 8) var(x1 x2 Ly)
(obs = 155)

Iteration 0: residual SS = 65.09422
Iteration 1: residual SS = 17.74304
Iteration 2: residual SS = 17.74304
Iteration 3: residual SS = 17.74304

Nonlinear regression Number of obs = 155
HAC kernel (lags): Newey-West (8) R-squared = 0.9732
Adj R-squared = 0.9727
Root MSE = .342788
Res. dev. = 103.919

------------------------------------------------------------------------ ------
| HAC
y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
------------- +----------------------------------------------------------------
/a0 | .9031516 .0344509 26.22 0.000 . 8350835 .9712196
/b0 | -.1591713 1.405989 -0.11 0.910 -2.937123 2.61878
/b1 | 1.929905 .428637 4.50 0.000 1.083004 2.776805
/b2 | .9768766 .5850655 1.67 0.097 -. 1790952 2.132848
------------------------------------------------------------------------ ------
Parameter b0 taken as constant term in model

. eststo NLreg

. esttab NLreg, p nostar ar2 aic bic mlabels(,nodepvar) eqlabels (,none) modelwidth(16)

-----------------------------
(1)
NLreg
-----------------------------
a0 0.903
(0.000)

b0 -0.159
(0.910)

b1 1.930
(0.000)

b2 0.977
(0.097)
-----------------------------
N 155
adj. R-sq 0.973
AIC 111.9
BIC 124.1
-----------------------------
p-values in parentheses


// Attempt #1: Completely separate -estimates- sets result

. nlcom (d1: _b[/b0]/(1-_b[/a0]) - (1-_b[/b1])*2) (d2: (_b[/b0]/(1-_b [/a0]) - 2)/(1-_b[/b1])), post

d1: _b[/b0]/(1-_b[/a0]) - (1-_b[/b1])*2
d2: (_b[/b0]/(1-_b[/a0]) - 2)/(1-_b[/b1])

------------------------------------------------------------------------ ------
y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
------------- +----------------------------------------------------------------
d1 | .2163 14.06365 0.02 0.988 -27.57064 28.00324
d2 | 3.918154 14.28661 0.27 0.784 -24.30932 32.14563
------------------------------------------------------------------------ ------

. estadd scalar d1 = _b[d1]

. estadd scalar d2 = _b[d2]

. esttab NLreg, p nostar ar2 aic bic mlabels(,nodepvar) eqlabels (,none) modelwidth(16) stats(d1 d2)

-----------------------------
(1)
NLreg
-----------------------------
a0 0.903
(0.000)

b0 -0.159
(0.910)

b1 1.930
(0.000)

b2 0.977
(0.097)
-----------------------------
d1
d2
-----------------------------
p-values in parentheses

. esttab ., p nostar ar2 aic bic mlabels(,nodepvar) eqlabels(,none) modelwidth(16)

-----------------------------
(1)

-----------------------------
d1 0.216
(0.988)

d2 3.918
(0.784)
-----------------------------
N 155
adj. R-sq
AIC .
BIC .
-----------------------------
p-values in parentheses


// Attempt #2: Clumsy; no SEs / p-values in table

. nlcom (d1: _b[/b0]/(1-_b[/a0]) - (1-_b[/b1])*2) (d2: (_b[/b0]/(1-_b [/a0]) - 2)/(1-_b[/b1]))

d1: _b[/b0]/(1-_b[/a0]) - (1-_b[/b1])*2
d2: (_b[/b0]/(1-_b[/a0]) - 2)/(1-_b[/b1])

------------------------------------------------------------------------ ------
y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
------------- +----------------------------------------------------------------
d1 | .2163 14.06365 0.02 0.988 -27.57064 28.00324
d2 | 3.918154 14.28661 0.27 0.784 -24.30932 32.14563
------------------------------------------------------------------------ ------

. mat d = r(b)

. estadd scalar d1=d[1,1]

. estadd scalar d2=d[1,2]

. esttab NLreg, p nostar ar2 aic bic mlabels(,nodepvar) eqlabels (,none) modelwidth(16) stats(d1 d2)

-----------------------------
(1)
NLreg
-----------------------------
a0 0.903
(0.000)

b0 -0.159
(0.910)

b1 1.930
(0.000)

b2 0.977
(0.097)
-----------------------------
d1 0.216
d2 3.918
-----------------------------
p-values in parentheses


What I ultimately would like is something akin to the following table, which I have constructed by hand via copy-and-paste; I'd like to be able to automate construction of such a table in a .do file (ultimately to produce tables in SMCL and LaTeX formats):

-----------------------------
(1)
NLreg
-----------------------------
a0 0.903
(0.000)

b0 -0.159
(0.910)

b1 1.930
(0.000)

b2 0.977
(0.097)
-----------------------------
d1 0.216
(0.988)

d2 3.918
(0.784)
-----------------------------
N 155
adj. R-sq 0.973
AIC 111.9
BIC 124.1
-----------------------------
p-values in parentheses


Thanks again for any assistance.

-- Mike

*
* 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