Analyze the relationship of selected US sectors with WTI oil before, during and after a crisis

Hello guys,

hope there is an econometrics pro who can give us some Feedback on our Project. My group and me want to analyze the relationship of WTI oil with selected US sectors before, during and after a crisis.

We use daily data from Datastream. Our time intervals are splitted in three periods:

• Period before the crisis. We decided to take the 12 months before the crisis.
• Period during the crisis. The length of the crisis itself.
• Period after the crisis is another 12 months Long.

We didnt find autocorrelation in the WTI oil residuals. Additionally we transformed the data to Returns.

We want to do the following with the data and the periods:

• For each period (Before, during, after crisis) we run separate regresssions for each sector separately. Additionally we run regressions with WTI oil as dependent and each single sector as Independent and the other way round (WTI oil indpendent and the sector as dependent). We take into account some lags from the dependent variable in our Regression. With this process we want to say whether the Independent granger causes the dependent.
• Additionally we run an ARMAX to quantify. So we write an ARMA with all the exogenous variables (X) as regressors.

Does this make sense, or do we miss an important fact?

Best Florian