[gov2001-l] Checking results in Part (b) - Gov2001 - lists.gking.harvard.edu

List overview All Threads
Download

[gov2001-l] Checking results in Part (b)

[gov2001-l] Confidence intervals...

[gov2001-l] Someone in your class...

acolin＠fas.harvard.edu

17 Mar 2008 17 Mar '08

7:15 p.m.

Hi all, Part (b) of the problem set asks to "use Zelig to check our results." Does "results" refer only to the vector of parameter estimates or should we also check standard errors? If the latter is true, then where does Zelig store the standard errors? (Sorry for not being able to find the answer in the docs.) Thank you! -Alexei

Reply

Show replies by date

yinanzhu＠fas.harvard.edu

17 Mar 17 Mar

7:17 p.m.

yo, what paper are you doing? And for once, i am starting the pset today!! On Mon, Mar 17, 2008 at 7:15 PM, Alexei Colin <acolin at fas.harvard.edu> wrote:

Hi all, Part (b) of the problem set asks to "use Zelig to check our results." Does "results" refer only to the vector of parameter estimates or should we also check standard errors? If the latter is true, then where does Zelig store the standard errors? (Sorry for not being able to find the answer in the docs.) Thank you! -Alexei _______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

-- Yinan Zhu Harvard Class of 2011 (626) 321-5878

Reply

acolin＠fas.harvard.edu

7:46 p.m.

Nevermind! Section code had the answer! vcov() is a magic function that extracts the var-co-var matrix from a fitted model object. Who knows where it is stored in the object... On 03/17/2008 07:15 PM, Alexei Colin wrote:

Hi all, Part (b) of the problem set asks to "use Zelig to check our results." Does "results" refer only to the vector of parameter estimates or should we also check standard errors? If the latter is true, then where does Zelig store the standard errors? (Sorry for not being able to find the answer in the docs.) Thank you! -Alexei _______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

Reply

jhainmueller＠gmail.com

8:44 p.m.

Yes vcov() works. Also check:

names(z.out1)

[1] "coefficients" "residuals" "fitted.values" "effects" "R" "rank" [7] "qr" "family" "linear.predictors" "deviance" "aic" "null.deviance" [13] "iter" "weights" "prior.weights" "df.residual" "df.null" "y" [19] "converged" "boundary" "call" "formula" "terms" "data" [25] "offset" "control" "method" "contrasts" "xlevels"

names(summary(z.out1))

[1] "call" "terms" "family" "deviance" "aic" "contrasts" "df.residual" [8] "null.deviance" "df.null" "iter" "deviance.resid" "coefficients" "aliased" "dispersion" [15] "df" "cov.unscaled" "cov.scaled" For various other stored objects. These can be accessed using for example: summary(z.out1)$cov.scaled etc. hope this helps, jens

-----Original Message----- From: gov2001-l-bounces at lists.fas.harvard.edu [mailto:gov2001-l- bounces at lists.fas.harvard.edu] On Behalf Of Alexei Colin Sent: Monday, March 17, 2008 7:46 PM To: gov2001-l at lists.fas.harvard.edu Subject: Re: [gov2001-l] Checking results in Part (b) Nevermind! Section code had the answer! vcov() is a magic function that extracts the var-co-var matrix from a fitted model object. Who knows where it is stored in the object... On 03/17/2008 07:15 PM, Alexei Colin wrote:

Hi all, Part (b) of the problem set asks to "use Zelig to check our results." Does "results" refer only to the vector of parameter estimates or should we also check standard errors? If the latter is true, then where does Zelig store the standard errors? (Sorry for not being able to find the answer in the docs.) Thank you! -Alexei _______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

_______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

Reply

acolin＠fas.harvard.edu

8:50 p.m.

Thanks, Jens! That explains the magic away. I did look at names(z.out), but didn't look at names(summary(z.out)). -alexei On 03/17/2008 08:44 PM, Jens Hainmueller wrote:

Yes vcov() works. Also check:

names(z.out1)

[1] "coefficients" "residuals" "fitted.values" "effects" "R" "rank" [7] "qr" "family" "linear.predictors" "deviance" "aic" "null.deviance" [13] "iter" "weights" "prior.weights" "df.residual" "df.null" "y" [19] "converged" "boundary" "call" "formula" "terms" "data" [25] "offset" "control" "method" "contrasts" "xlevels"

names(summary(z.out1))

[1] "call" "terms" "family" "deviance" "aic" "contrasts" "df.residual" [8] "null.deviance" "df.null" "iter" "deviance.resid" "coefficients" "aliased" "dispersion" [15] "df" "cov.unscaled" "cov.scaled" For various other stored objects. These can be accessed using for example: summary(z.out1)$cov.scaled etc. hope this helps, jens

-----Original Message----- From: gov2001-l-bounces at lists.fas.harvard.edu [mailto:gov2001-l- bounces at lists.fas.harvard.edu] On Behalf Of Alexei Colin Sent: Monday, March 17, 2008 7:46 PM To: gov2001-l at lists.fas.harvard.edu Subject: Re: [gov2001-l] Checking results in Part (b) Nevermind! Section code had the answer! vcov() is a magic function that extracts the var-co-var matrix from a fitted model object. Who knows where it is stored in the object... On 03/17/2008 07:15 PM, Alexei Colin wrote:

Hi all, Part (b) of the problem set asks to "use Zelig to check our results." Does "results" refer only to the vector of parameter estimates or should we also check standard errors? If the latter is true, then where does Zelig store the standard errors? (Sorry for not being able to find the answer in the docs.) Thank you! -Alexei _______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

_______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

_______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

Reply

5901

days inactive

5902

days old

gov2001@lists.gking.harvard.edu

Manage subscription

4 comments

3 participants

tags (0)

participants (3)

acolin＠fas.harvard.edu
jhainmueller＠gmail.com
yinanzhu＠fas.harvard.edu