Threshold, Gaps Not allowed

[Mina Sami]

I am working with threshold regression: I have quarterly data (strongly balanced)
I have Panel Data strongly balanced with quarterly data over the period 2000q1-2017q2 and around 200 group
Stock_return=constant+ B interest_rate

I got always an error: gaps not allowed

When I run:
threshold stock_return , threshvar(interest_rate)

I got error:
gaps not allowed

Any way to deal with this?

Please note I tried to use : xthreg command. It worked! I have significant threshold . The problem : F-test for the whole model is not significant, which means I should avoid Fixed effects!!

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output and sample data using dataex. Being able to replicate your problem is extremely helpful to us helping you.

It is not clear that threshold works for panel data. It is in the time series documentation and many time series models won't work for panels. You can check this by experimenting with artificial data you create.

On xtreg, that the F for the whole model is not significant does not imply no fixed effects. There is a separate F for the u_i which is the one that tells you about using panel effects. The model F is for the included variables and insignificant F there means your variable is not explaining the dv..

[Phil Bromiley]

see also user written xthreg.

[Mina Sami]

Dear Phil, Thank you for your reply. My variables are significant but F for the u_i is not significant. So I should drop the model?