I am considering extending a multi-factor fixed income stochastic model (e.g. LIBOR-Market) to use stochastic correlation matrices instead of determinstic ones.
For pricing instruments with short maturities stochastic correlation would not really make sense - for the correlation structure is in general much more stable than the volaltility. Thus in the short run it is often sufficient to work with a determinstic correlation.
Instruments with long maturities (as often encountered in fixed income markets) are a different matter entirely.
- Are there precedents of such models (either equity or fixed income)?
- Whould the added value be worth the effort ? (or is it enough to deal with stochastic volatility which is certainly more important)