Journal Search Help

markontap

Hi guys,
I was hoping someone could point me in the direction of some journal entries or working papers that deal specifically with time varying risk of security market returns? I've had some success but I'm sure there must be more out there more closely related to the topic. I've searched through Jstor and Repec along with other mentioned in the sticky at the top of the page...

I appreciate any help!

Économiste Monétaire

Have you tried looking through papers by Engle, Campbell, Shiller, Fama, French, etc.? Are you attempting to estimate a (G)ARCH-M model?

markontap

EM,

Forgot to reply to the thread.. Thanks for the info; I was on the right track, just wanted to make sure I wasn't leaving anything out. I'm doing an undergrad 'thesis' on volatility in returns. The econometrics I'm doing ends at multivariate regressions and only touches on heteroskedasticity. Are ARCH models and the appliction of them covered in any books or will I have to work off the published journal papers? Engle, Bollerslev etc

Thanks again,
M

Économiste Monétaire

Did you cover any time series in your econometrics classes? Particularly stationarity and ARIMA models? ARCH models are simple enough to understand. Take your standard linear regression model

[latex] \displaystyle y = x' \beta + \epsilon [/latex]

which you've seen before, but add in something about the disturbance term, [latex] \displaystyle \epsilon [/latex], namely

[latex] \displaystyle \epsilon_{t} = v_{t} \sqrt{h_{t}} [/latex]

where [latex] \displaystyle h_{t} = \alpha_{0} + \sum_{j=1}^{q} \alpha_{j} \epsilon_{t-j}^{2} [/latex]

This is an ARCH(q) model. It's a bit different from the heteroskedasticity you've come across in that the disturbance term is conditionally heteroskedastic w.r.t. it's past values, rather than, say, the explanatory variables, x'. You can extend this to GARCH(p,q), which is just adding an autoregressive term of order p to 'h' above.

In terms of books, most (modern) intermediate to advanced econometrics textbooks will have a section on conditional heteroskedasticity. It's a popular topic in finance when building up to option pricing theory, so books like Analysis of Financial Time Series, by Ruey Tsay, cover them in reasonable detail. You don't need to read the technical papers, but I would glance through the Engle at al. ARCH-M paper if you're looking at time-varying risk premium in excess returns.

Do you have any particular series in mind?