Hello,
When analyzing the performance of a strategy with a rolling beta, is there a way to detect if the beta is increasing (in a statistical way)?
For example say you have an algorithm that is trying to profit from the momentum factor in equities, in a market neutral fashion. You get a low beta (to SPX/MSCI ACWI/whatever) that you are happy with. However, a graph of the rolling beta (say a 12 month window) gives a different picture - it appears that equity beta is growing over time, and it hits levels that you would normally not be happy with.
Is there a way to test the rolling beta and say whether or not the increase in beta is statistically significant? -- a method with some statistical rigor beyond saying, "yep, that beta sure does look like it's gone up".
Thank you!
Phill