Variance Ratio Test - Exponential Random Walk

[doug_johnson7]

Hi,

I am conducting a variance ratio test on a certain time period of returns for the S&P 500 and am a bit confused on how to properly execute the exponential random walk test (Log SP500). I have conducted the random walk test using the closing prices of the S&P, but I am not sure if for the exponential random walk test if I should be using the daily returns that can be calculate by subtracting today's price from yesterday's price in excel prior to inputting the data into EViews or if EViews will difference the data itself given the closing prices. Simply put, I am wondering if I should be using the daily closing prices and just switching the test to an exponential random walk, or if I need to alter the data I am inputting into the test to be the calculated returns rather than the closing prices.

Just as an additional note, when I conducted the exponential test using the returns calculated in excel, the probability comes out as 0.000 for the joint test and for the majority of the individual tests for each of the user-specified lags (2, 4, 8, 16, 32, etc.).

Thanks for the help!
Doug