VAR Granger Casuality Test for FDI,Production and TFP HELP!

[lau]

Hi All,

CAN PLEASE SOMEONE HELP ME!?!!? Im doing my dissertation ( due in December )and I want to analyze the effect of FDI on the country productivy and on TFP. I want to do a VAR to test for Granger casuality, between FDI, production, and TFP and do a comparative analysis between two countries (USA and Mexico). I think I have all the data but I DONT KNOW WHAT THE HECK TO DO !?!?!!? I've never do a VAR in my life and Im not very familiar with time series. I wanted to know if you can help me organizing this mess. I have the data in excel but I dont even know how to get it into Eviews. I know.... I know... my life sucks! but can you please help me to see if I can seee the light at the end of the tunnel?!?!?!

Thanks so much!!!! :D
Lau

[loooooo]

Hi Lau,

You should first read any econometrics textbook coz how you should go about it depends on data and sometimes it's very hard to make a judgement call without having a firm foundation on econometrics, but I can at least outline what you should be doing:
1. check if your variables are I(1) series - ADF test, PP test, even panel unit root testing, and many more. I suspect FDI and Production may be I(1) but I don't know what TFP is.
2. On the condition that all three are I(1), you can check for cointegration.
3. If at least one of them is I(0), you should set up a VAR model - whether to difference all of them or only those that are I(1) should depend on the literature (if you have a clear guideline derived from your hypothesis, etc. on this issue, do as it suggests).
4. Please note granger causality test is only for short-term dynamics and doesn't take into account contemporaneous relations but only lags.