Illegal date

[nsea411]

Dear friends,

I encountered a problem when I tried to run a program( I wrote this code because I want to cut a whole time series into several samples according to the date, for example, the data from 2001/1/1-2001/1/31 is a sample, the data from 2001/2/1-2001/2/28 is another sample, and then do regressions within each sample).

Code: Select all

  wfopen C:\Users\dell\Desktop\agriculture.xlsx

for !i= 1 to 12
for !j=2001 to 2019

smpl {!j}m{!i} {!j}m{!i}
genr a{!j}{!i} =ri
genr b{!j}{!i} =roil
genr c{!j}{!i} =rm
genr d{!j}{!i} =smb
genr e{!j}{!i} =hml

matrix(5,(!i-1)*19+!j-2000) coefs

equation eq{!j}{!i}.ls a{!j}{!i} c b{!j}{!i} c{!j}{!i} d{!j}{!i} e{!j}{!i}

colplace(coefs, eq{!j}{!i}.@coefs,(!i-1)*19+!j-2000)
next
next

When I run this code, Eviews gave me the notice: Error in Sample: Illegal date 1/1/2001 in "SMPL 2001M1 2001M1".
I wonder what goes wrong .
If I can get your help, I will be very happy. Thank you in advance！

[EViews Gareth]

Can you provide the Excel file?

[nsea411]

Dear Gareth,
I have posted two Excel files, agriculture data are calculated by the previous code. When I tried to run it again, the code works, but when I wanted to run a similar code(attached below) to calculate coal data, the same Eviews notice came out again. I'm using Eviews 10(I'm sorry I forgot to mention it before).
Thank you for your help!

Code: Select all

wfopen C:\Users\dell\Desktop\coal.xlsx

for !i= 1 to 12
for !j=2003 to 2020

smpl {!j}m{!i} {!j}m{!i}

genr a{!j}{!i} =ri
genr b{!j}{!i} =roil
genr c{!j}{!i} =rmrf
genr d{!j}{!i} =smb
genr e{!j}{!i} =hml

matrix(5,(!i-1)*18+!j-2000) coefs

equation eq{!j}{!i}.ls a{!j}{!i} c b{!j}{!i} c{!j}{!i} d{!j}{!i} e{!j}{!i}

colplace(coefs, eq{!j}{!i}.@coefs,(!i-1)*18+!j-2000)
next
next

[EViews Gareth]

You can only set a sample using dates if the workfile is dated.

[nsea411]

Sorry Mr. Gareth, I don't quite understand what you mean. Could you please be more specific?
Thank you very much!

[EViews Gareth]

Your excel file is not dated

[startz]

The problem is that a bunch of your date series is missing values.