turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Data Management
- /
- Forum
- /
- Panel Data Regressions

Topic Options

- RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

01-03-2015 02:01 AM

Hi Folks

i have panel data. Cross sectional Variable=id, Time series variable is t, when i try to run the panel regression using the following statement;

proc sort data=bank.mergid1; by id t; run;

proc panel data=mergid1;

id id t;

model ROAA=MS CR NIM SIZE EQTA OHTA OBSTA LG DC STTR/pooled;

run;

**then following lines appear in log **

370 proc panel data=bank.mergid1;

371 id id t;

372 model ROAA=MS CR NIM SIZE EQTA OHTA OBSTA LG DC STTR/pooled;

373 run;

ERROR: There is only one cross section or time series observation. Computations will be terminated.

NOTE: The SAS System stopped processing this step because of errors.

NOTE: PROCEDURE PANEL used (Total process time):

real time 0.32 seconds

cpu time 0.32 seconds

**Note: my sas data file is attached **

Regards

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to raqthesolid

01-04-2015 08:57 AM

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to raqthesolid

01-04-2015 09:32 AM

I only know how to do it in REGRESSION.

proc reg;

mode y=x1 x2... /** vif** ;

run;

check variance inflated factor , if variables'vif gt 10 then it has muliti-linear correlation .

or you can use proc corr to plot it correlated matrix graph and check their muliti-linear correlation .

I suggest you post it at Steve might give you right answer.

or

SAS Forecasting and Econometrics

since it is a statistical problem and it is about economic model .

Xia Keshan

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Ksharp

01-04-2015 09:49 AM

Thanks man but i want to create an ID variable from bank column ??how to do this ...

Regards

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to raqthesolid

01-04-2015 01:45 PM

Post your log with the error message.

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Reeza

01-06-2015 09:52 AM

hi

when i use the following codes;

proc sort data=mergid1; by id t; run;

proc panel data=mergid1;

id id t;

model ROAA=MS CR NIM SIZE EQTA OHTA OBSTA LG DC STTR/pooled;

run;

then these are the line that appear when i try to run the Panel data regression.

370 proc panel data=bank.mergid1;

371 id id t;

372 model ROAA=MS CR NIM SIZE EQTA OHTA OBSTA LG DC STTR/pooled;

373 run;

ERROR: There is only one cross section or time series observation. Computations will be terminated.

NOTE: The SAS System stopped processing this step because of errors.

NOTE: PROCEDURE PANEL used (Total process time):

real time 0.32 seconds

cpu time 0.32 seconds

kindly help me to solve this issue. thanks

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to raqthesolid

01-06-2015 10:09 AM

Not familiar with PROC PANEL but the message seems be saying that you do not have multiple time values. So the value of the variable T is constant.

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

01-06-2015 10:51 AM

how can i make T as time series i tried everything i can but still it gives this message i uploaded a new data set. i will be very thankful if you can help me

Regards

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to raqthesolid

01-06-2015 06:51 PM

1. Eliminate countries (ids) where you have only one observation (one time period).

2. Then eliminate duplicate id t combinations.

proc sort data=bank.mergid1 out=mergid1 NOUNIQUEKEY; by id; run;

proc sort data=mergid1 out=mergid1 NODUPKEY; by id year; run;

proc panel data=mergid1;

...

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to raqthesolid

01-04-2015 03:34 PM

So the variables you listed in your MODEL statement are not in the dataset you posted.