[Davorin Jersic]

Dear Statalist users,

I am using Stata 12 and I am just learning how to use it.

I have a problem following the instructions on how to fix the "selectindex" problem (I have read the post that has also the procedure on how to deal with the problem, but I can't fix it).

Can please somebody help me? Maybe provide more detailed instructions, because I am lost.

Thank you very much for your help.

Davorin

[Kendal zhang]

Dear Majid,

I am glad you liked the "Log of Gravity". About your questions:

1. Indeed your models appear to be failing the RESET. First of all, please check that you are performing the test correctly; our webpage has code illustrating how to do it. If the model really fails the test, you need to think about ways of improving your model. Maybe you can add other important regressors? Or maybe you just need to include interactions (or cross-products) of the regressors that you already have.

2. The main reason to prefer PPML is not the zeros but the heteroskedasticity of trade data. So, even without zeros, PPML is generally preferable.

Finally, note that your sample is quite small, so try to keep your model as parsimonious as possible.

Best regards,

Joao

Dear Joao,

I have a question about the gravity model. I want to analyze the impact of political conflict on bilateral trade just between China and USA, while many articles are involved in the analysis of the influencing factors of multinational trade volume. I wonder if this model is applicable to the analysis of two countries.

Beat wishes,

Kendal

[Joao Santos Silva]

Dear Kendal zhang,

I am not sure to fully understand your question, but I guess that there are at least two possible replies. Shocks to trade between China and the US will impact other trade flows (for example, less trade between China and the US may lead to more trade between China and Canada) and the MLR terms should capture that. So, the model with MLR estimated with data for all countries is interesting even if you only care about a particular pair. If you only plan to use data on trade on China and the US, you can continue to include the MLR terms, but they will not play the same role and it is up to you to decide whether you want to keep them in and only you know what you want to estimate.

Best wishes,

Joao

[Kendal zhang]

Dear Joao,

Yes, that is what my meaning. But if I estimate the data for all countries with Gravity model, how can I know influence degree of shocks to the trade between China and USA? Thank you.

Best wishes,

Kendal

[Joao Santos Silva]

Dear Kendal zhang,

Finding out how to do that is your contribution, but you can perhaps use dummies for that particular pair.

Best wishes,

Joao