Statalist


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

RE: Rép. : st: RE: Exporting results after -anova- [wi[UTF-8] th -estout-?]


From   "Martin Weiss" <[email protected]>
To   <[email protected]>
Subject   RE: Rép. : st: RE: Exporting results after -anova- [wi[UTF-8] th -estout-?]
Date   Wed, 10 Dec 2008 15:01:38 +0100

Line for the server...

This eerie feeling creeps up on me that all of this could be done in two lines in MATA, but I have no book that teaches me how...

HTH
Martin


-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Herve STOLOWY
Sent: Wednesday, December 10, 2008 3:02 PM
To: [email protected]
Subject: Rép. : st: RE: Exporting results after -anova- [wi[UTF-8] th -estout-?]

Dear Martin:

Thanks again for your help. I have everything now.

local BIGIF if round>7 & round<26 & win==1 & offer==1
anova return info risk info*risk `BIGIF'
matrix c1 =  e(mss), e(df_m), e(mss)/e(df_m) , e(F),
1-F(e(df_m),e(df_r),e(F))
forv i=1/3{
matrix c2_`i' = e(ss_`i'), e(df_`i'), e(ss_`i')/ e(df_`i'), e(F_`i'),
1-F(e(df_`i'),e(df_r),e(F_`i'))
}
matrix c3 = c1 \ c2_1  \ c2_2  \ c2_3 
matrix c4 = e(rss), e(df_r), e(rss)/e(df_r), ., .
matrix c5 = c3 \ c4
matrix rownames c5 = "Model" "Info" "Risk" "Info*Risk" "Residual"
matrix colnames c5 = "Partial SS" "df" "MS" "F" "p-value"
matrix list c5
mat2txt, matrix(c5) saving(anova) title(Panel B: Return, disclosure and
risk) replace

Output
matrix list c5

c5[5,5]
           Partial SS          df          MS           F     p-value
    Model   .06789462           3   .02263154   32.137905           0
     Info   .00734185           1   .00734185   10.425786   .00126014
     Risk   .04520712           1   .04520712   64.196351   1.776e-15
Info*Risk   .01024621           1   .01024621   14.550124   .00014007
 Residual   1.6231828        2305    .0007042           .           .

Best regards

Hervé

***********************************************************
Professeur/Professor
President of the French Accounting Association (AFC)
HEC Paris
Departement Comptabilite Controle de gestion / Dept of Accounting and
Management Control
1, rue de la Liberation
78351 - Jouy-en-Josas
France
Tel: +33 1 39 67 94 42 - Fax: +33 1 39 67 70 86
mail: stolowy at hec dot fr
web: http://www.hec.fr/stolowy
>>> "Martin Weiss" <[email protected]> 10/12/08 14:06 >>>
Line for the server...

This has been a German/French collaboration, I guess :-)
Anyway, just divide the SS by the df for the MS.

************
webuse systolic
anova systolic drug disease drug*disease
di in red "MSS Model: " %11.6fc e(mss)/e(df_m)
************


HTH
Martin


-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Herve STOLOWY
Sent: Wednesday, December 10, 2008 8:40 AM
To: [email protected]
Subject: [UTF-8] Rép. : Re: Rép. : RE: Rép.[UTF-8] : st: RE:
Exporting
results after -anova- [wi[UTF-8] th -estout-?]

Dear Martin:

Thank you very much for the syntax. With your help, I have managed to
get almost everything:

anova return info risk info*risk if round>7 & round<26 & offer==1 &
win==1
matrix c1 =  e(mss), e(df_m), . , e(F), 1-F(e(df_m),e(df_r),e(F))
matrix list c1
forv i=1/3{
matrix c2_`i' = e(ss_`i'), e(df_`i'), . , e(F_`i'),
1-F(e(df_`i'),e(df_r),e(F_`i'))
}
matrix c3 = c1 \ c2_1  \ c2_2  \ c2_3 
matrix c4 = e(rss), e(df_r), ., ., .
matrix c5 = c3 \ c4
matrix rownames c5 = "Model" "Info" "Risk" "Info*Risk" "Residual"
matrix colnames c5 = "Partial SS" "df" "MS" "F" "p-value"
matrix list c5
mat2txt, matrix(c5) saving(anova) title(Panel B: Return, disclosure and
risk) replace

Here is the output: 


 matrix list c5

c5[5,5]
           Partial SS          df         MS           F     p-value
    Model   .06789462           3           .   32.137905           0
     Info   .00734185           1           .   10.425786   .00126014
     Risk   .04520712           1           .   64.196351   1.776e-15
Info*Risk   .01024621           1           .   14.550124   .00014007
 Residual   1.6231828        2305           .           .           .


The only elements I don't know how to get them are the MS.

Best regards

Herve



***********************************************************
Professeur/Professor
President of the French Accounting Association (AFC)
HEC Paris
Departement Comptabilite Controle de gestion / Dept of Accounting and
Management Control
1, rue de la Liberation
78351 - Jouy-en-Josas
France
Tel: +33 1 39 67 94 42 - Fax: +33 1 39 67 70 86
mail: stolowy at hec dot fr
web: http://www.hec.fr/stolowy
>>> "Martin Weiss" <[email protected]> 09/12/08 23:26 >>>
Herve,

what is it exactly terms of the example I provide. Getting the p-values is easy, as you can

see. Does anyone know how I could specify the upper bound of the -forv-
so 
that it would work with any number of covariates? I cannot find a
returned 
value that would give me this number...

*******
webuse systolic, clear
anova systolic drug disease drug* disease
di in red "Model: " %5.4fc 1-F(e(df_m),e(df_r),e(F))
forv i=1/3{
di in red "Covariate `i': " %5.4fc 1-F(e(df_`i'),e(df_r),e(F_`i'))
}
*******

HTH
Martin
_______________________
----- Original Message ----- 
From: "Herve STOLOWY" <[email protected]>
To: <[email protected]>
Sent: Tuesday, December 09, 2008 10:14 PM
Subject: Rép. : RE: Rép. : st: RE: Exporting results after -anova- 
[with -estout-?]


> Dear Martin:
>
> Thank you for the tip. I am a great fan of -estout- and have used it a
> lot. For the moment, I don't see how it could work. I know that I can
> get a regression with anova, regression. I have done this but I still
> need the p-values for my variables: risk, info and risk*info (the ones
> displayed after anova).
>
> Best regards
>
> Herve
>
> ***********************************************************
> Professeur/Professor
> President of the French Accounting Association (AFC)
> HEC Paris
> Departement Comptabilite Controle de gestion / Dept of Accounting and
> Management Control
> 1, rue de la Liberation
> 78351 - Jouy-en-Josas
> France
> Tel: +33 1 39 67 94 42 - Fax: +33 1 39 67 70 86
> mail: stolowy at hec dot fr
> web: http://www.hec.fr/stolowy
>>>> "Martin Weiss" <[email protected]> 09/12/08 14:19 >>>
> Line for the server...
>
> Type -ssc hot- to see the candidates for your task. -outreg2- and
> -estout-
> can do it. My attempts at a solution for you are currently choking on
> the
> fact that -e(b)- after -anova- is not part of what you want to see in
> your
> output file. Rather, you want all of those ereturns like - e(mss)- in
> there... I will let you know when I got it :-)
>
> HTH
> Martin
>
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Herve
STOLOWY
> Sent: Tuesday, December 09, 2008 2:06 P> Thank you for your reply. I
looked at the tabletutorial file but did
not
> find anything specific on -anova-. Should I use the saved results and
> matrices?
>
> Best regards
>
> Herve
>
>
> ***********************************************************
> Professeur/Professor
> President of the French Accounting Association (AFC)
> HEC Paris
> Departement Comptabilite Controle de gestion / Dept of Accounting and
> Management Control
> 1, rue de la Liberation
> 78351 - Jouy-en-Josas
> France
> Tel: +33 1 39 67 94 42 - Fax: +33 1 39 67 70 86
> mail: stolowy at hec dot fr
> web: http://www.hec.fr/stolowy
>>>> "Martin Weiss" <[email protected]> 09/12/08 13:29 >>>
> Line for the server...
>
> You might want to consult - ssc d tabletutorial-. Is your -if-
qualifier
> material to the problem?
>
> HTH
> Martin
>
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Herve
STOLOWY
> Sent: Tuesday, December 09, 2008 1:21 PM
> To: [email protected]
> Subject: st: Exporting results after -anova-
>
> Dear Statalisters:
>
> I run the following anova command:
>
> anova return info risk info*risk if round>7 & round<26 & offer==1 &
> win==1
>
> Would you know a way to export the output (or part of the output) into
a
> tab-delimited file?
>
> Best regards
>
> Herve Stolowy
>
> ***********************************************************
> Professeur/Professor
> President of the French Accounting Association (AFC)
> HEC Paris
> Departement Comptabilite Controle de gestion / Dept of Accounting and
> Management Control
> 1, rue de la Liberation
> 78351 - Jouy-en-Josas
> France
> Tel: +33 1 39 67 94 42 - Fax: +33 1 39 67 70 86
> mail: stolowy at hec dot fr
> *   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/
>
>
> *
> *   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/
>
> *
> *   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/



*
*   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/

*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index