Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: RE: RE: RE: st: RE: extract t-values


From   "Martin Weiss" <martin.weiss1@gmx.de>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: RE: RE: RE: st: RE: extract t-values
Date   Fri, 7 Nov 2008 17:39:55 +0100

Note that most replies point toward -parmest-, Bastian, so the case for
doing this by hand is getting weaker by the minute. Still, it can make sense
to go back to the roots and observe where all the stuff that fancy packages
do automatically actually comes from...


HTH
Martin


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Bastian Steingros
Sent: Friday, November 07, 2008 5:28 PM
To: statalist@hsphsun2.harvard.edu
Subject: Re: RE: RE: RE: st: RE: extract t-values

"So e(df_r) is available after -reg, robust-, right?"

Obviously, it works. I compared my self-made results with the regression
output and the p-values are totally identical. 
(I do not know why that e(df_r) did not work before.)

Do you think it should not be working? Using e(df) does not create any
p_values.

By the way, I have a question to e(.) in general .
How can I put digits captured by the command e(.)  into my dataset?
I mean, can I create for example : gen degree_freedom=e(df_r) ? If so, how?

B


-------- Original-Nachricht --------
> Datum: Fri, 7 Nov 2008 13:02:59 +0100
> Von: "Martin Weiss" <martin.weiss1@gmx.de>
> An: statalist@hsphsun2.harvard.edu
> Betreff: RE: RE: RE: st: RE: extract t-values

> So e(df_r) is available after -reg, robust-, right?
> 
> 
> HTH
> Martin
> 
> 
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Bastian
> Steingros
> Sent: Friday, November 07, 2008 12:44 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: RE: RE: st: RE: extract t-values
> 
> Hi Martin,
> voicing concern might be logical when looking at my descriptions of the
> problem. I think I should have detailed it. 
> However, the command you last emailed is exactly what I searched!!!
> 
> Many thanks!
> 
> have a nice day
> B
> 
> 
> 
> 
> 
> -------- Original-Nachricht --------
> > Datum: Fri, 7 Nov 2008 12:20:04 +0100
> > Von: "Martin Weiss" <martin.weiss1@gmx.de>
> > An: statalist@hsphsun2.harvard.edu
> > Betreff: RE: RE: st: RE: extract t-values
> 
> > I tried to make my example as general as possible, hence the
> "complicated"
> > syntax and the use of M.Buis` -ssc d indeplist-. You can shrink this
> > syntax
> > if all you want is a calculation for one of the covariates, i.e.
> > 
> > **************
> > sysuse auto, clear
> > reg p w f r, robust
> > gen tweight = _b[weight]/_se[weight]
> > gen pweight = 2*ttail(e(df_r),abs(_b[weight]/_se[weight]))
> > **************
> > 
> > I do not endorse your strategy, and others have voiced concerns as well,
> > but
> > if you absolutely want it...
> > 
> > 
> > HTH
> > Martin
> > 
> > 
> > -----Original Message-----
> > From: owner-statalist@hsphsun2.harvard.edu
> > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Bastian
> > Steingros
> > Sent: Friday, November 07, 2008 12:12 PM
> > To: statalist@hsphsun2.harvard.edu
> > Subject: Re: RE: st: RE: extract t-values
> > 
> > Yes I have.
> > 
> > I think using this command could be helpful:
> > di in red "p-value: " %5.4fc
> > 2*ttail(e(df_r),abs(_b["`indep'"]/_se["`indep'"])) _n(2)
> > 
> > But, I do not fully understand how to incorporate this command in my
> > dofile.
> > 
> > I would like to program *gen p_value=...*. Is something of that
> mentioned
> > above fit to do so?
> > I have the t-values for each year. 
> > I guess it must be *gen p_value=2*ttail(e(df_r), abs(t_value))... but I
> > then
> > have a syntax problem to deal with... Maybe e(df_r) does not function
> when
> > using reg ... , robust
> > 
> > Or is it not possible to calculate the p-values by hand in stata ? 
> > 
> > Hope I made my problem  clear
> > 
> > B
> > 
> > 
> > -------- Original-Nachricht --------
> > > Datum: Fri, 7 Nov 2008 11:11:17 +0100
> > > Von: "Martin Weiss" <martin.weiss1@gmx.de>
> > > An: statalist@hsphsun2.harvard.edu
> > > Betreff: RE: st: RE: extract t-values
> > 
> > > Are you sure you have read through all messages in this thread? You
> > might
> > > have missed
> > > 
> > > http://www.stata.com/statalist/archive/2008-11/msg00234.html
> > > 
> > > 
> > > 
> > > HTH
> > > Martin
> > > 
> > > -----Original Message-----
> > > From: owner-statalist@hsphsun2.harvard.edu
> > > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Bastian
> > > Steingros
> > > Sent: Friday, November 07, 2008 11:09 AM
> > > To: statalist@hsphsun2.harvard.edu
> > > Subject: Re: st: RE: extract t-values
> > > 
> > > Hi Martin hi everybody
> > > 
> > > you are right, parmest simplifies it greatly, but is there another way
> > to
> > > put the p-values directly into my data set?
> > > I mean, using something like gen t=_b[.]/_se[.] makes stata write the
> > > t-statistics in the *same* dataset where the other variables already
> > are. 
> > > The thing is I run many regressions. My dataset consists of data from
> > 1950
> > > to 1995. I want to run regression  for 5-year-time-windows only, that
> > is,
> > > the first time window is 1950-1955 etc. 
> > > To put all the statistics  from the regression table into my dataset
> > > simplifies my task because I then can reduce my sample to those time
> > > windows
> > > where all the regression coefficient are at least significant at the
> 5%
> > > level (for example by using: bys window: keep if p_value=<0.05).
> > > 
> > > That is the reason why I want to calculate the t-values and, in the
> next
> > > step, the p-values separately. 
> > > 
> > > Is there a may to calculate p-values after having calculated t-values?
> > > 
> > > Try to give an example of how it looks like in my dataset
> > > year   var1 var2 ...  coefficient         t_val         p_val
> > > 
> > > 1950                          2.98             4.27   -->  ????       
>  
> > > 1950                          2.98             4.27   ---> ????
> > > ...etc.
> > > 
> > > Basti
> > > -------- Original-Nachricht --------
> > > > Datum: Thu, 6 Nov 2008 17:56:48 +0100
> > > > Von: "Martin Weiss" <martin.weiss1@gmx.de>
> > > > An: statalist@hsphsun2.harvard.edu
> > > > Betreff: st: RE: extract t-values
> > > 
> > > > BTW, your -gen- statement is probably not the most expedient way to
> > > > achieve
> > > > what I assume you want. You might want to look into -local-s or
> > > -scalar-s
> > > > for more efficient storage. Or indeed -ssc d parmest- which
> simplifies
> > > > this
> > > > task greatly...
> > > > 
> > > > 
> > > > HTH
> > > > Martin
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: owner-statalist@hsphsun2.harvard.edu
> > > > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Bastian
> > > > Steingros
> > > > Sent: Thursday, November 06, 2008 5:48 PM
> > > > To: statalist@hsphsun2.harvard.edu
> > > > Subject: st: extract t-values
> > > > 
> > > > Hi Stata users,
> > > > 
> > > > does anybody know a ansver to this problem:
> > > > 
> > > > I want to extract the t-values from the table displayed below. To
> > obtain
> > > > the
> > > > coefficients is easy by programming gen coef1=_b[var1]
> > > > 
> > > > Is there also a command to extract the t-value of var1???
> > > > 
> > > >
> > >
> >
>
----------------------------------------------------------------------------
> > > > --
> > > > dep_var |      Coef.       Std. Err.       t         P>|t|      
> [95%
> > > > Conf.
> > > > Interval]
> > > >
> > >
> >
>
---------+------------------------------------------------------------------
> > > > --
> > > >    var1   |  -.0074074   .0008079     -9.168   0.000      -.0090039
> > > > -.0058109
> > > >    _cons |   4.711111   .2162561     21.785   0.000       4.283786
> > > > 5.138436
> > > >
> > >
> >
>
----------------------------------------------------------------------------
> > > > --
> > > > 
> > > > thanks 
> > > > 
> > > > B
> > > > -- 
> > > > "Feel free" - 10 GB Mailbox, 100 FreeSMS/Monat ...
> > > > Jetzt GMX TopMail testen: http://www.gmx.net/de/go/topmail
> > > > *
> > > > *   For searches and help try:
> > > > *   http://www.stata.com/help.cgi?search
> > > > *   http://www.stata.com/support/statalist/faq
> > > > *   http://www.ats.ucla.edu/stat/stata/
> > > > 
> > > > *
> > > > *   For searches and help try:
> > > > *   http://www.stata.com/help.cgi?search
> > > > *   http://www.stata.com/support/statalist/faq
> > > > *   http://www.ats.ucla.edu/stat/stata/
> > > 
> > > -- 
> > > "Feel free" - 10 GB Mailbox, 100 FreeSMS/Monat ...
> > > Jetzt GMX TopMail testen: http://www.gmx.net/de/go/topmail
> > > *
> > > *   For searches and help try:
> > > *   http://www.stata.com/help.cgi?search
> > > *   http://www.stata.com/support/statalist/faq
> > > *   http://www.ats.ucla.edu/stat/stata/
> > > 
> > > *
> > > *   For searches and help try:
> > > *   http://www.stata.com/help.cgi?search
> > > *   http://www.stata.com/support/statalist/faq
> > > *   http://www.ats.ucla.edu/stat/stata/
> > 
> > -- 
> > GMX Download-Spiele: Preizsturz! Alle Puzzle-Spiele Deluxe über 60%
> > billiger.
> >
>
http://games.entertainment.gmx.net/de/entertainment/games/download/puzzle/in
> > dex.html
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/help.cgi?search
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> > 
> > 
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/help.cgi?search
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> 
> -- 
> Psssst! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen:
> http://www.gmx.net/de/go/multimessenger
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
> 
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

-- 
"Feel free" - 10 GB Mailbox, 100 FreeSMS/Monat ...
Jetzt GMX TopMail testen: http://www.gmx.net/de/go/topmail
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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