Tag: correlation (Page 2 of 4)
Estimating from historical data requires many assumptions about similarity. Reducing the number of estimated parameters can control model risk.
Correlation is often used to proxy diversification, but correlations can be very unstable over time. It's akin buying insurance that may or may not pay out.
Historically, fixed income has provided diversification to equities. What happens if the correlation of fixed income to equities inverts?
Having a risk metric that is never surprising can be dangerous. If your risk measure does not surprise you, then the market will.
In this post, we look at a new measure of oscillatory similarity that agrees more with our intuition of what "correlation" means.
