topic Re: How can I cluster standard erros in proc panel in SAS Forecasting and Econometrics

How can I cluster standard erros in proc panel

SayeE — Wed, 25 Sep 2019 22:36:30 GMT

I have the following code for my two-way fixed effect model. In my dataset "data1", there are a few "units" in each "firm". How I can get the heteroskadasticity-robust standard errors clustered by "firm".

proc sort data=data1;

by unit time;

run;

proc panel data=data1;

id unit time;
model DV = x /FIXTWO;
run;

Re: How can I cluster standard erros in proc panel

xyz0505 — Fri, 27 Sep 2019 14:05:13 GMT

Hi,

Right now PROC PANEL only provides cluster s.e by id variable.

Re: How can I cluster standard erros in proc panel

xyz0505 — Fri, 06 Mar 2020 16:16:53 GMT

There's a trick you can use to cluster on variable that is not ID using proc panel. If you have an ID variable (maybe firm), time variable, and an ID2 variable you want to cluster on (let's say industry, for example).

First sort the data according to id2 variable, and create a new variable called time2. time2 variable give sequence for each observation in each id2 group. Example looks like:

id	t	id2	t2
1	1	1	1
1	2	1	2
1	3	1	3
1	4	1	4
1	5	1	5
1	6	1	6
2	1	1	7
2	2	1	8
2	3	1	9
2	4	1	10
2	5	1	11
2	6	1	12
3	1	2	1
3	2	2	2
3	3	2	3
3	4	2	4
3	5	2	5
3	6	2	6
4	1	2	7
4	2	2	8
4	3	2	9
4	4	2	10
4	5	2	11
4	6	2	12

Then sort the data according to id2 and time2.

Then run:

proc panel data = panel_example3 ;
   Id id2 t2;
     class id;
     model y = x1 x2 id / pooled hccme=0 cluster;
run;

should give you fixone model with fixed effects on id but clustering on id2.

id	t	id2	t2
1	1	1	1
1	2	1	2
1	3	1	3
1	4	1	4
1	5	1	5
1	6	1	6
2	1	1	7
2	2	1	8
2	3	1	9
2	4	1	10
2	5	1	11
2	6	1	12
3	1	2	1
3	2	2	2
3	3	2	3
3	4	2	4
3	5	2	5
3	6	2	6
4	1	2	7
4	2	2	8
4	3	2	9
4	4	2	10
4	5	2	11
4	6	2	12

id	t	id2	t2
1	1	1	1
1	2	1	2
1	3	1	3
1	4	1	4
1	5	1	5
1	6	1	6
2	1	1	7
2	2	1	8
2	3	1	9
2	4	1	10
2	5	1	11
2	6	1	12
3	1	2	1
3	2	2	2
3	3	2	3
3	4	2	4
3	5	2	5
3	6	2	6
4	1	2	7
4	2	2	8
4	3	2	9
4	4	2	10
4	5	2	11
4	6	2	12

id	t	id2	t2
1	1	1	1
1	2	1	2
1	3	1	3
1	4	1	4
1	5	1	5
1	6	1	6
2	1	1	7
2	2	1	8
2	3	1	9
2	4	1	10
2	5	1	11
2	6	1	12
3	1	2	1
3	2	2	2
3	3	2	3
3	4	2	4
3	5	2	5
3	6	2	6
4	1	2	7
4	2	2	8
4	3	2	9
4	4	2	10
4	5	2	11
4	6	2	12