topic Re: Standard deviation in panel data in SAS Procedures

Standard deviation in panel data

namrata — Mon, 23 Jul 2012 05:25:28 GMT

Hello

Could anybody tell me how to compute std dev of a variable in panel data?Is this right?

data want;

set have;

by gvkey year;

vol=std(variable);

run;

I tried this but the error is:

The STD function call does not have enough arguments.

thanks.

Re: Standard deviation in panel data

Ksharp — Mon, 23 Jul 2012 05:37:11 GMT

Or SQL ?

proc sql;

create table want as

select gvkey,year,std(variable) as std

from have

group by gvkey,year

;

quit;

Ksharp

Re: Standard deviation in panel data

ArtC — Mon, 23 Jul 2012 05:42:10 GMT

The SUMMARY procedure, among others, will also provide univariate summary statistics.

proc summary data=have;

class gvkey year;

var variable;

output out=summry /autoname;

run;

Re: Standard deviation in panel data

Ksharp — Mon, 23 Jul 2012 08:58:48 GMT

Arthur.Carpenter,

Long time no see.

Maybe you miss NWAY option ,since OP don't want all levels of class variables.

Xia Keshan

Re: Standard deviation in panel data

ArtC — Mon, 23 Jul 2012 18:33:02 GMT

Thanks Ksharp - good catch. Or the WAYS statement which offers more flexibility.

Re: Standard deviation in panel data

namrata — Mon, 23 Jul 2012 22:20:45 GMT

Thanks Ksharp & ArtC.

I tried the code:

proc sql;

create table want as;

select gvkey, year, std(ni) as std

from final1

group by gvkey,year;

quit;

The std column remains blank,

Re: Standard deviation in panel data

Reeza — Mon, 23 Jul 2012 22:22:25 GMT

you have an extra semi colon after the as in your first line after proc sql.

Check your log.

Re: Standard deviation in panel data

namrata — Mon, 23 Jul 2012 22:34:25 GMT

Sorry Reeza...that was a typo.I was not being able to copy my code directly from SAS.

There is no semi colon in that line.The table want is created with gvkeys & years...however std values are missing.

Re: Standard deviation in panel data

ArtC — Mon, 23 Jul 2012 22:43:07 GMT

Silly question but can you verify that there are at least two values of NI for each combination of your GROUPBY variables, and that NI is not missing for at least two obs within the BY group?

Re: Standard deviation in panel data

namrata — Mon, 23 Jul 2012 22:52:32 GMT

Yes ,there are atleast 2 obs against each id.

Re: Standard deviation in panel data

namrata — Tue, 24 Jul 2012 00:13:20 GMT

I need to change my requirement a little..Want the std dev upto the previous year,essentially the lag std dev. Could anybody please help me with this?

Id year ni std

1 2000 5 -

1 2001 1 std(ni upto 2000)

1 2002 2 std(ni upto 2001)

1 2003 3 std(ni upto 2002)

2 2001 2 -

2 2002 5 std(ni upto 2001)

2 2003 5 std(ni upto 2002)

2 2004 1 std(ni upto 2003)

3 2002 6

3 2003 6

3 2004 4

Re: Standard deviation in panel data

ArtC — Tue, 24 Jul 2012 05:08:30 GMT

Are you still getting missing values for the STD? My test code gives the expected results.

Re: Standard deviation in panel data

ArtC — Tue, 24 Jul 2012 05:45:45 GMT

In your revised problem I will assume that the column NI is the mean of the NI values. This problem is a form of a running average problem. I will tackle it here using a multlabel format - there are other approaches. In the following code substitute SEX AGE and WEIGHT for your variables the approach will be the same.

proc sort data=sashelp.class out=have;

by sex age;

run;

proc print data=have; run;

proc format ;

value yrgrp (multilabel)

10 = '10'

10-11 = '10-11'

10-12 = '10-12'

10-13 = '10-13'

10-14 = '10-14'

10-15 = '10-15'

10-16 = '10-16'

10-17 = '10-17'

10-18 = '10-18';

run;

proc summary data=have;

class sex;

class age/mlf;

var weight;

ways 2;

output out=groups n= mean= std=/autoname;

format age yrgrp.;

run;

proc print data=groups; run;

Re: Standard deviation in panel data

namrata — Thu, 26 Jul 2012 19:55:44 GMT

ArtC, I am so sorry for this late reply. I have not been able to try out the suggestion since I have a submission soon and I decided to try out the code for the extensive paper that I shall work on after the submission.

I apologize once again & thank you so much for writing out the code for me.I shall definitely try it out and get back to you.

Re: Standard deviation in panel data

ArtC — Thu, 26 Jul 2012 20:07:19 GMT

Let us know how it works for you.