Quantitative finance collector
C++ Matlab VBA/Excel Java Mathematica R/Splus Net Code Site Other
Dec 19

How to Combine Long and Short Return Histories Efficiently

Posted by abiao at 11:06 | Paper Review | Comments(0) | Reads(7291)
Missing data imputation is a common technique many researchers have to apply for some certain situations, especially when we do some portfolio analysis that requires an equal length of historical returns of assets in the portfolio. Typically we assume a distribution of the underlying data and simulate missing data based on the assumption, MLE or EM algorithm is used for simulation. For example, a great R package I have introduced for missing data imputation was at here.

"How to Combine Long and Short Return Histories Efficiently" is a good paper forthcoming in Financial Analysts Journal by Sébastien Page, as introduced
A common challenge in portfolio risk analysis is that certain assets have shorter return histories than others. Unfortunately, many standard portfolio risk analysis techniques—including historical tail risk measurement, regime-dependent risk analysis, and bootstrapping simulations—require full return histories for all assets or risk factors. The author presents easy instructions on how to efficiently combine data for investments whose histories differ in length and offers a new model to better account for non-normal distributions.

An important feature of this paper is instead of assuming that the uncertainty around the backfilled returns is normally distributed, the model samples empirical residuals from the short sample. Evidence shows this method is efficient. The author also provides Matlab code in the Appendix for us to play around.


Add a comment
Enable HTML
Enable UBB
Enable Emots
Nickname   Password   Optional
Site URI   Email   [Register]