0
$\begingroup$

I am testing the cointegration between these variables:

Gold Price (Ringgit),
Exchange Rate - MYR to USD (Ringgit),
Real Effective FX Rate Based on CPI,
T-Bill 10 Years Rate,
Consumer Price Index.

Which variables should I take logarithms of before running the tests?

Improve this question
$\endgroup$
1
  • $\begingroup$ Gold Price (Ringgit), Exchange Rate - MYR to USD (Ringgit), Real Effective FX Rate Based on CPI, T-Bill 10 Years Rate, & Consumer Price Index $\endgroup$
    – sidewinder2203
    Commented Dec 17, 2015 at 7:00

1 Answer 1

Reset to default
0
$\begingroup$

In general, Exchange rate is logged, CPI is logged, Gold Price is logged. Interest rates are left like this.

I would suggest, however, to apply the following transformations, because people usually use those: CPI becomes inflation and exchange rate becomes a ap-/depreciation rate.

Improve this answer
$\endgroup$
5
  • $\begingroup$ So why not we log for interest rate? And then i am actually taking all the data in differenced form for all the tests (stationary, johansson & engel granger, var). Is it the correct way? $\endgroup$
    – sidewinder2203
    Commented Dec 17, 2015 at 16:01
  • $\begingroup$ Because it makes no sense to log an interest rate. How would you interpret it ? You take the log of a series because you suspect an underlying exponential form. This is not the case for interest rates. $\endgroup$
    – Olivier Hubert
    Commented Dec 18, 2015 at 9:00
  • $\begingroup$ I have another question. Say for differenced formation. Do i do the first difference for all variables or it depends on the nature of data. For eg. consumer price index? $\endgroup$
    – sidewinder2203
    Commented Dec 18, 2015 at 9:03
  • $\begingroup$ Your aim is to have stationary variables when estimating. Thus, those which are stationary should not be differenced. $\endgroup$
    – Olivier Hubert
    Commented Dec 18, 2015 at 9:06
  • $\begingroup$ So we should just go for stationarity test first before deciding on the differencing formation? $\endgroup$
    – sidewinder2203
    Commented Dec 18, 2015 at 9:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.