vecrank error message r498 the sample has gaps

[Asha Lata]

Dear all,

I am currently analysing a panel of 5 countries and 7 variables over a period of 39 years, ie 195 observations overall. The panel is strongly balanced. Having tsset my data, I attempted to perform Johansen's Cointegration Test using vecrank on Stata-12 and received the message r(498) the sample has gaps. Please refer to the attached photo. Using tsreport, report report0 list, I understood the gaps were due to panel changes. So once I typed tsreport, report report0 list panel, the results indicated the sample had no gaps.

My mistake is that I assumed the problem resolved so I again ran the vecrank on all 7 variables. And once again received the same error message r(498) the sample has gaps.

I would greatly appreciate any assistance in solving this gap business so I can proceed with this cointegration test.

Thanks and regards,
Asha

[Nandit Paul]

Hi Asha, I was also performing the same command but my data finds the gaps if I type in tsreport. But I don't know how to fill in the gaps. I want to run the commands.

tsreport

Time variable: dateval
---------------------------
Starting period = 31dec2010
Ending period = 01oct2014
Observations = 979
Number of gaps = 196

[rizwan ahmad]

same problem with me, is there someone who solve the problem

[rizwan ahmad]

this happems with me , nandit paul and asha lata is there any solution to this problem

[Anat Tchetchik]

Hi Asha Lata, Nandit Paul , rizwan ahmad,
I had the same problem yesterday. I think that the problem is that -vecrank- does not work with panel. you can either loop the command on each panel unit separably or try the -xtwest- (ssc install) which is anError–correction–based cointegration tests for panel data http://www.stata-journal.com/article...article=st0146 hope it helps
(btw I could not "get rid" of the bold format in the text above- sorry for that)
Anat

[Lanie Santos]

Hi! I also encountered the same problem when I ran:
vecrank CO2 FDI, trend(constant)
the sample has gaps
r(498);

But my data is strongly balanced:
tsset Country Year, yearly
panel variable: Country (strongly balanced)
time variable: Year, 1991 to 2014
delta: 1 year

However, when i ran tsreport , detail, the result shows as follows:

Observations with preceding time gaps
(gaps include panel changes)
----------------------------------
Record | Country Year
----------+-----------------------
25 | SGP 1991
49 | THA 1991
----------------------------------

i have three countries being studied with complete data from 1991 to 2014 for each. how can i resolve this issue? I need to run cointegration test in my study. please advice. if you can please provide me the script, i would appreciate it so much.

[Abdulhadi Aliyara]

vecrank GDPper L_p K_P fdi_p , trend(constant) max
the sample has gaps
please any solution

[Quynh-Anh Thi Nguyen]

Hi! I also encountered the same problem when I ran:
vecrank CO2 FDI, trend(constant)
the sample has gaps
r(498);

But my data is strongly balanced:
tsset Country Year, yearly
panel variable: Country (strongly balanced)
time variable: Year, 1991 to 2014
delta: 1 year

However, when i ran tsreport , detail, the result shows as follows:

Observations with preceding time gaps
(gaps include panel changes)
----------------------------------
Record | Country Year
----------+-----------------------
25 | SGP 1991
49 | THA 1991
----------------------------------

i have three countries being studied with complete data from 1991 to 2014 for each. how can i resolve this issue? I need to run cointegration test in my study. please advice. if you can please provide me the script, i would appreciate it so much.

Dear,

Did you solve the problem?