Errors in xthreg

[Jojo Melchi]

Good morning everybody.
I am trying to detect threshold in panel regression model using xthreg command. But there are still error messages. Can anybody help me to deal with this?

 xthreg DZ L.( gdp snp shiller_hp) s1-s3, rx(L.change_fedrate) qx(L.tgap6) thnum(1) trim(0.05) bs(300) nobslog r
> obust
Estimating  the  threshold  parameters:   1st ......                 thest():  3200  conformability error
                thestm():     -  function returned error
                 <istmt>:     -  function returned error
r(3200);

[Jojo Melchi]

I do really need help! I don't know where the problem stems from.

[Phil Bromiley]

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.

This is a user written routine - it could easily have some characteristics that differ from most Stata routines. With conformability errors, you might have a variable being dropped due to colinearity - what happens if you run a very simple model? If that doesn't fit it, you can try to figure out the problem by looking closely at the ado file. That is not easy sometimes. Or, you can ask the authors.

[Jojo Melchi]

Thank you Phil,
I tried different trimming percentage. It works for trimming value greater than 5%. I wonder if this is still correct.
For exemple:

 
 xthreg DZ L.( gdp snp shiller_hp) s1-s3, rx(L.change_fedrate) qx(L.tgap5) thnum(1) trim(0.06) bs(300) nobslog robust

or

 
 xthreg DZ L.( gdp snp shiller_hp) s1-s3, rx(L.change_fedrate) qx(L.tgap6) thnum(1) trim(0.3) bs(300) nobslog robust

I wonder if the analysis and my estimators are still unbiased.
Thank you for anyexplanation

[Jojo Melchi]

Hello everybody, you can find the data in attachement. The individual variable is "no" and the time variable is date.
Any help is welcome

[Jojo Melchi]

Is there anybody who can help me with this issue?

[River Huang]

Hi, Jojo: I think that a 10% (but not 30%) trimming is still acceptable!

[Jojo Melchi]

Thank you River!

[Abdessamad Raghibi]

Jojo Melchi I too have the same problem but even if I change the trimming to 30% (which is too much) I still get the r(3200) error.
Here is my code:

xthreg ROA Cash Size_TA MB LEV, rx(Cash) qx(Cash) thnum(1) trim(0.1) grid(400) bs(300)

[Shah Parth]

Hello Everyone,

I am new to the Stata and know a little bit about xthreg command

Could someone explain to me about rx(varlist) and qx(varname) what are these two variables to be mentioned in STATA?

[River Huang]

Shah Parth Please help xthreg.

Options

    rx(varlist) is the regime-dependent variable.  Time-series operators are allowed.  rx() is required.

    qx(varname) is the threshold variable.  Time-series operators are allowed.  qx() is required.