Page 1 of 1

trend stationary and difference stationary

Posted: Mon May 10, 2010 8:00 am
by shevchenko62
I have few 4 time series data: stock price, M2, CPI and interest rates.
M2, CPI and interest rates are all I(1).
As for stock price, it is non-stationary, but it seems like having deterministic trend. After detrending and running ADF on level with intercept on residual data, it is showing stationarity. My questions are
1) Is stock price considered I(1)? Am I right to run ADF on level with intercept only on the residual after detrending? What is the significance of the result of ADF on level which is showing stationarity?
2) can M2, CPI, interest rate and stock price be examined using cointegration?

Re: trend stationary and difference stationary

Posted: Mon May 10, 2010 9:13 am
by startz
Stock price is the classic
example of an I(1) series

Re: trend stationary and difference stationary

Posted: Mon May 10, 2010 4:15 pm
by shevchenko62
The stock price that I use is a finance index after applying natural logarithmic. I just found out that M2 also has similar attributes whereby it seems to have deterministic trend. It is non stationary at level. But showing some up trend. After detrending, I run ADF (intercept only) and it is showing stationarity now. How can I proceed with cointegration or can I proceed with cointegration actually?



I suppose I need to look at the trend coefficient (@TREND)to determine if it is significant to determine if it has deterministic trend right? what is the value of @TREND that is deemed significant?