Hello,
I have to test for structural breaks in a time series; in particular, I want first to check for breaks and then to check if there is a break in 1992. I found that I should use the command estat sbknown, so my code is:
tsset country year
xi: reg score i.year i.country
breakvars(year)
And then
estat sbknown, break(1992)
However, when I try to run it, Stata gives me the error message “sample may not include multiple panels-r(459)”. I don’t understand why it says “multiple panels”; I only have one dataset where the same countries are followed over time. I thought the problem could arise when using tsset, which gives me the output:
tsset country year
panel variable: country (unbalanced)
time variable: year, 1957 to 2019, but with gaps
delta: 1 unit
But I doubt it.
Does anyone know why I get the error?
Thanks everyone in advance!
I have to test for structural breaks in a time series; in particular, I want first to check for breaks and then to check if there is a break in 1992. I found that I should use the command estat sbknown, so my code is:
tsset country year
xi: reg score i.year i.country
breakvars(year)
And then
estat sbknown, break(1992)
However, when I try to run it, Stata gives me the error message “sample may not include multiple panels-r(459)”. I don’t understand why it says “multiple panels”; I only have one dataset where the same countries are followed over time. I thought the problem could arise when using tsset, which gives me the output:
tsset country year
panel variable: country (unbalanced)
time variable: year, 1957 to 2019, but with gaps
delta: 1 unit
But I doubt it.
Does anyone know why I get the error?
Thanks everyone in advance!
Comment