返回列表 发帖

Bonds, MBS and std deviation

What's the deal with using std deviation as a measure of risk for bonds and MBS?

I thought that std dev was no good for bonds, but I'm sure I saw a question somewhere which said that you could use std dev for bonds. Am I missing something?

i dont recall seeing using std deviation as measure of risk for bonds due to fact that:

1) calculation of std dev becomes too complex (requires many asumptions of variances and covariance) as number of bonds in the portfolio increases

2) bond characterstics change as time to maturity lessens. so if i calculate std dev based on charactersitcs at t=0 for a 5 yr duration bond, its std dev will be a lot different than what i calculate at time t = 1 or any other time. the two wont be comparable.

TOP

Also returns for bonds are negatively skewed (can lose everything but you never get more). Another reason you can't use standard deviation.

TOP

Not in the curriculum. Focus on the LOSes people!!! Whilst this stuff is interesting, study this stuff on 7th June!

TOP

返回列表