Volume & Volatility

[keshav.jayant]

Dear All,

I am trying to demonstrate the relationship between volume of trades and stock return volatility. I am using EViews6 for this purpose. My question is:
How do I include the varibale log(volume) in the mean equation of my GARCH(1,1) process?
I tried the following equation in EViews but I am not sure if it is correct or not and hence require help on this. The equation I entered is as follows:

returns=c(1) + c(2)*Log(Volume)

I entered the above equation in the equation box and then clicked ARCH process and it gave me the estimation output however I am not entirely sure if this is correct. If it is, does it satisfy the equation: (PLEASE SEE ATTACHED WORD FILE)

I really require this help urgently!! thanks everyone!

[trubador]

It seems the volume of trades is an explanatory variable for the variance equation only. However, you specified it in the mean equation. You should enter the variable log(volume) into the "Variance Regressors" section of the "Equation Estimation" dialog box.

[keshav.jayant]

Hi trubador,

Thanks for your reply. If you can see the attached file, it is the OUTPUT I got after running the GARCH(1,1) model.
Is it correct to enter volume in both the mean equation and variance equation? or only the variance equation alone?

Please see the attached file and let me know. Thank you.

[trubador]

Statistically speaking, volume variable seems to have significant impacts on both mean and volatility of returns. I cannot say whether this is correct or not, but I would also try GARCH-in-mean specification while keeping the volume variable in the mean equation in order to control for the mean-variance relationship.

[keshav.jayant]

Alright ill keep that mind. thanks alot!