[Sarah S.]

Hi,

I am running xthenreg but it generates an error ' estimates post: matrix has missing values'. I have checked many times and there is no missing value. Any advice please?

Thanks
Sarah

[zaki tchy]

hello dear, I am wondering if you get some useful information about xthenreg. I still don't know what is the difference between b and d in the results, and which one is above and under the threshold. Please any answer!!!
Thank you.

[Alina Mulyukova]

Hi Sarah,

I also receive an error saying "estimates post: matrix has missing values" but i don't have any missing values in the data. Did you figure out how to deal with it?

Best,
Alina

[Zarifah Fasihah]

Hi everyone.

Is anyone get how to interpret the output as in #1?
sedki zn did you already figure out how to interpret that?


Regard,
Zarifah

[Zarifah Fasihah]

Dear all

Finally we get the stata' command to deal with the endogeneity issue for threshold model, but it still not clear.

i mean how can we know if it exists the threshold effect and at which level




[ATTACH=CONFIG]n1495809[/ATTACH]




so please how to deal with the outputs using " xthenreg "
kind regards
Sedki,

Hi everyone.

Is anyone get how to interpret the output as in #1?
sedki zn did you already figure out how to interpret that?


Regard,
Zarifah

[Syed Sadaqat Shah]

Dear sedki zn,
I got some comments from reviewers about the threshold model that I estimated. I followed xtendothresdpd command in Stata 16 to estimate threshold level for public debt. The specific comments are as follow;

1. If a theoretical model is proposed there will be opportunity to consider the issue of reverse causation. In other words, countries that have high growth potential may be able to sustain higher debt levels and hence can sell more debt. Thus cross-country differences in thresholds may reflect that future expected growth impacts on the ability to service larger debt volumes as well as other factors that are identified in the paper. This concept is alluded to in discussing why inflation as a threshold is worth investigation.
2. Perhaps a multiple threshold model is required to pick up these different impacts i.e. use both inflation and debt as a threshold?.
3. Why do you divide the dataset a priori. Could you estimate the whole dataset and then introduce a dummy variable which would separate out the two group to see if the dummy variable is significant? In this context would all explanatory variables be the same across the two groups?
4. It is clear that you can identify the non-linear relationship between public debt and economic growth by using dynamic panel threshold model. However, it is not clear whether you have solved the endogeneity problem successfully.
5. Estimate the effects of the control variables under different debt regimes. You find single threshold value for the debt but we know that the effects of control variables vary under different debt regimes such lower debt regime and higher debt regimes. That is, what are the effects of the control variables on economic growth (GDP PC Capita growth rate) in case of full sample, in lower debt regime and higher debt regime.

Thanks.

[Wang Yue]

Hi, dear. Do you know how to use stata to test the threshold model, or how to get sup W of the bootstrap linearity test? Please.

[Obioma Asuzu]

Please how do we fix the error message that says "estimates post: matrix has missing values"? Thank you

[Manh Hoang Ba]

Dear sedki zn ,

When you use the threshold model, your sample will be diveded into two regimes (below and under the threshold level). Which means that you have to take both of variables (x_b and x_d). But i don't know if b means below and d means under or not .

Is your threshold variable endogenous? Because you didn't mention your instruments in the regression.

best regards,
Fairouz

Hi fai ot ,
x_b is the beta parameter and x_d is the delta parameter in the theoretical model of Seo & Shin (2016, DOI: 10.1177/1536867X19874243)

[Manh Hoang Ba]

Please how do we fix the error message that says "estimates post: matrix has missing values"? Thank you

Hi Sarah S., Alina Mulyukova and Obioma Asuzu,
It looks like the number of diagonal units in your data is less than the integer in grid_num(integer). The default value is 100, so if your number of diagonal units is less than 100, try declaring a smaller value, for example 50.