ARCH (autoregressive conditional heteroscedasticity) is a statistical model that considers the variance of the current error term to be a function of the variances of the previous time periods’ error terms. I heard that this model made Prof. Engle a Nobel prize recipient.

Heteroscedesticity in regression problems in astronomy has been discussed with various kind of observations since astronomers are interested in correlations and causality between variables in the data set. Perhaps, the heteroscedasticity in volatilty of finance does not have any commonality with astronomical times series data, which could be the primary reason this celebrated model does not appear in ADS. However, there are various times series models branched from ARCH that consider heteroscedasticity in errors and I hope some can be useful to astronomers for analyzing times series data with inhomogeneous errors.

[Note] All links lead to wikipedia.org

Leave a comment