Dissertation question - correlation/cointegration

[adampc]

Dear all,

for my thesis I am looking at the relationship between an equity index and a few other indices which are not equity based. I intend to look at the correlation between them which is fine however, I am not sure how to go about looking into cointegration as I am fairly new to stats and am doing this study upon reccomendation from my supervisor.

I understand that I need to make sure the data is stationary and to look at the t-1 residual plots to do this. I know this is probably an extremely basic question, but could somebody please explain the steps in lay-mans terms for me?

Just to clarify, I have quarterly returns for each index over a 25 year period and also have the rebased data started at 100 for each index. So in addition to carrying out cointegration on the data set, I would like to know whether you think I should regress the quarterly returns, the rebased index, or a log of the index?

I really appreciate any help :-)

Thanks,

Adam

[trubador]

Searching the forum may be helpful. For instance, CF Tang (aka tcfoon) has made useful and detailed explanations on cointegration from both theoretical and practical point of view. Here is the link: http://forums.eviews.com/viewtopic.php? ... &sk=t&sd=a