Reading DATA!!!

[celineboulenger13]

Hi, I am new at RATS (I have version 8 ) and I have been having problems loading my panel data from excel. I have the set up the booklet recommends (having the data set up per individual) and I have tried having the names of the series on the first row and then not having the names but every time I get the same error, here is the code I am using:

OPEN DATA "C:\Documents\paneldataforstata.xlsx"
CALENDAR(PANELOBS=23,A) 1990:1
DATA(FORMAT=XLSX,ORG=COLUMNS) 1//1990:01 552//2012:01 individual year gdp cap labor re

and here is the error I am getting:

## IO30. There is no series INDIVIDUAL on the file
## IO30. There is no series YEAR on the file
## IO30. There is no series GDP on the file
## IO30. There is no series CAP on the file
## IO30. There is no series LABOR on the file
## IO30. There is no series RE on the file

My data looks like :

Code: Select all

country year	gdp	            capital	        labor 	re
1	1990	4.27251E+11	97675046357	8502257	14.756
1	1991	4.25431E+11	88491579483	8548010	16.433
1	1992	4.2715E+11	84954717509	8623259	15.8744
1	1993	4.44303E+11	90624439154	8675174	17.164
1	1994	4.62274E+11	95802590018	8832057	16.879
1	1995	4.80215E+11	1.06241E+11	9049690	16.469
1	1996	4.99171E+11	1.09159E+11	9154738	16.575
1	1997	5.18856E+11	1.16407E+11	9230553	17.733
1	1998	5.41845E+11	1.27666E+11	9331221	16.702
1	1999	5.68961E+11	1.33482E+11	9428920	17.701
1	2000	5.90991E+11	1.44166E+11	9624371	17.422
1	2001	6.02389E+11	1.32714E+11	9793326	18.001
1	2002	6.2566E+11	1.44637E+11	9937598	17.712
1	2003	6.44926E+11	1.62769E+11	10101919	18.509
1	2004	6.71733E+11	1.7665E+11	10229628	18.52
1	2005	6.93339E+11	1.87794E+11	10555735	19.95
1	2006	7.14074E+11	2.05223E+11	10806943	21.481
1	2007	7.40924E+11	2.15657E+11	10986204	21.066
1 	2008	7.68355E+11	2.3615E+11	11285373	19.61
1	2009	7.81662E+11	2.41189E+11	11522495	19.62
1	2010	7.97001E+11	2.46267E+11	11715010	22.351
1	2011	8.15503E+11	2.55332E+11	11882313	26.773
1	2012	8.45904E+11	2.83966E+11	12027327	25.14892
2	1990	2.23054E+11	50950902671	3524707	32.359
2	1991	2.30731E+11	55189848688	3576695	32.367
2	1992	2.35561E+11	55265428595	3668918	35.856
2	1993	2.36802E+11	54589018456	3705767	37.74
2	1994	2.4249E+11	57398066802	3825983	36.596
2	1995	2.4896E+11	57508229022	3849977	38.605
etc... 

For 24 countries total!

Could someone please help me out? I am getting very frustrated. It says in the booklet that RATS easily reads panel data so I must be missing something here.

Thanks in advance!

[TomDoan]

Hi, I am new at RATS (I have version 8 ) and I have been having problems loading my panel data from excel. I have the set up the booklet recommends (having the data set up per individual) and I have tried having the names of the series on the first row and then not having the names but every time I get the same error, here is the code I am using:

OPEN DATA "C:\Documents\paneldataforstata.xlsx"
CALENDAR(PANELOBS=23,A) 1990:1
DATA(FORMAT=XLSX,ORG=COLUMNS) 1//1990:01 552//2012:01 individual year gdp cap labor re

and here is the error I am getting:

## IO30. There is no series INDIVIDUAL on the file
## IO30. There is no series YEAR on the file
## IO30. There is no series GDP on the file
## IO30. There is no series CAP on the file
## IO30. There is no series LABOR on the file
## IO30. There is no series RE on the file

My data looks like :

Code: Select all

country year	gdp	            capital	        labor 	re
1	1990	4.27251E+11	97675046357	8502257	14.756
1	1991	4.25431E+11	88491579483	8548010	16.433
1	1992	4.2715E+11	84954717509	8623259	15.8744
1	1993	4.44303E+11	90624439154	8675174	17.164
1	1994	4.62274E+11	95802590018	8832057	16.879
1	1995	4.80215E+11	1.06241E+11	9049690	16.469
1	1996	4.99171E+11	1.09159E+11	9154738	16.575
1	1997	5.18856E+11	1.16407E+11	9230553	17.733
1	1998	5.41845E+11	1.27666E+11	9331221	16.702
1	1999	5.68961E+11	1.33482E+11	9428920	17.701
1	2000	5.90991E+11	1.44166E+11	9624371	17.422
1	2001	6.02389E+11	1.32714E+11	9793326	18.001
1	2002	6.2566E+11	1.44637E+11	9937598	17.712
1	2003	6.44926E+11	1.62769E+11	10101919	18.509
1	2004	6.71733E+11	1.7665E+11	10229628	18.52
1	2005	6.93339E+11	1.87794E+11	10555735	19.95
1	2006	7.14074E+11	2.05223E+11	10806943	21.481
1	2007	7.40924E+11	2.15657E+11	10986204	21.066
1 	2008	7.68355E+11	2.3615E+11	11285373	19.61
1	2009	7.81662E+11	2.41189E+11	11522495	19.62
1	2010	7.97001E+11	2.46267E+11	11715010	22.351
1	2011	8.15503E+11	2.55332E+11	11882313	26.773
1	2012	8.45904E+11	2.83966E+11	12027327	25.14892
2	1990	2.23054E+11	50950902671	3524707	32.359
2	1991	2.30731E+11	55189848688	3576695	32.367
2	1992	2.35561E+11	55265428595	3668918	35.856
2	1993	2.36802E+11	54589018456	3705767	37.74
2	1994	2.4249E+11	57398066802	3825983	36.596
2	1995	2.4896E+11	57508229022	3849977	38.605
etc... 

For 24 countries total!

Could someone please help me out? I am getting very frustrated. It says in the booklet that RATS easily reads panel data so I must be missing something here.

Thanks in advance!

Did you try using the Data Wizard? Obviously, one problem is that the series appears to be called "Country" on the file, not "Individual". If you can't read it using the Wizard, you would have to attach the actual data file.

[celineboulenger13]

Sorry I did change the series to be named individual and not country.

How do I use the data Wizard? Is that the "ratsdata" program? I did try using it but did not understand it. Sorry this is such an elementary question but I am very confused with RATS and how it works.

Thanks a lot for your help!

[TomDoan]

The Data Wizard is described beginning on page Int-18 of the Introduction. It's in RATS (not RATSDATA) on the Data/Graphics menu (Data (Other Formats) operation).

[celineboulenger13]

Yes I was using that but when I try to open the file that way nothing happens. And the series are not on the series log. Could you tell me the steps I am supposed to follow when I use the data wizard?

Thanks!!

[celineboulenger13]

So i tried using the Wizard again and this time it worked!
However I am working on doing panel FMOLS but I get error messages when I try to include a constant and when I use the series in logs instead of levels.

@panelfm(det=constant, average=simple, print=full)
# lgdp lcap llabor lre
## OP3. This Instruction Does Not Have An Option CON
## OP3. This Instruction Does Not Have An Option DET
## SR10. Missing Values And/Or SMPL Options Leave No Usable Data Points
The Error Occurred At Location 1801, Line 158 of PANELFM

Those are the error messages I get, they all go away if I use no constant and don't use the logs. However I should be able to do this with a constant and logs right?

Could you help with that?

Thanks a lot!

[TomDoan]

You'll have to post the whole program. Copy and paste it straight out of RATS.

[celineboulenger13]

OPEN DATA "C:\Users\Celine Boulenger\Documents\WinRATS 8\paneldataforstata.xlsx"
CALENDAR(PANELOBS=23,A) 1980:1
DATA(FORMAT=XLSX,NOLABELS,ORG=COLUMNS,TOP=2) 1//1980:01 24//2002:01 INDIVIDUAL YEAR GDP CAP LABOR RE
set lgdp = log(gdp)
set lcap = log(cap)
set llabor = log(labor)
set lre = log(re)
@panelfm(lags=3)
# lgdp lcap llabor lre
## SR10. Missing Values And/Or SMPL Options Leave No Usable Data Points
The Error Occurred At Location 1801, Line 158 of PANELFM
@panelfm(constant)
## OP3. This Instruction Does Not Have An Option CON

[TomDoan]

In the final one, DET=CONSTANT is the option, not CONSTANT. (And DET=CONSTANT is the default, so you don't need it anyway).

With the log data, have you checked your data to make sure you have (positive) data all the way through? (TABLE instruction?)

[celineboulenger13]

I understand for the constant but when I get results there is no constant coefficient. Do you know how to retrieve it?

For the logs, yes all of my numbers are positive and when I look at the log series they look totally fine; I really don't understand why it is not working.

Thanks,

[celineboulenger13]

It seems that @panelfm is only working for me when I use 3 variables in log and one in levels. But obviously that doesn't make sense to do. Do you have any idea what is going on?

Thanks,

[celineboulenger13]

Am I allowed to use @fm instead of @panelfm if I am trying to do Panel FMOLS?
When I use @fm I get rid of those "errors" I had with @panelfm. It lets me use the variables in natural logs. Please help me out! here is my result with @fm

@fm(lags=5)
# lgdp lcap llabor lre

Linear Regression - Estimation by Fully Modified LS
Dependent Variable LGDP
Panel(23) of Annual Data From 1//1981:01 To 88//1989:01
Usable Observations 551
Degrees of Freedom 547
Skipped/Missing (from 2010) 1459
Mean of Dependent Variable 26.857688906
Std Error of Dependent Variable 1.500298527
Standard Error of Estimate 0.188009839
Sum of Squared Residuals 19.335191729
Durbin-Watson Statistic 0.1915

Variable Coeff Std Error T-Stat Signif
************************************************************************************
1. Constant 3.2372185119 0.4985490730 6.49328 0.00000000
2. LCAP 0.9145869765 0.0446334757 20.49105 0.00000000
3. LLABOR 0.0274955653 0.0433791912 0.63384 0.52618376
4. LRE 0.0314858413 0.0104497359 3.01308 0.00258615

[TomDoan]

Is it possible that your data are constructed in a way that there is a linear combination of the four variables in logs with the individual-specific intercept is zero? That would produce the behavior that you describe. You can do @FM, but that assumes homogeneity in both the intercept and the cointegrating vector across individuals.

The deterministics are considered nuisance parameters in this model, so they aren't shown.