FI semivariance

Yes, contingent claim risk instead of call risk. How about the cap risk? That’s the one I am not so sure about. Where is CSK? emarkhans Wrote: ------------------------------------------------------- > Call risk wasn’t the choice. It was “contingent > claim” risk, which an MBS clearly is unless my > brain’s gone completely haywire.

so CSK and others… please u still think tht semi-variance is the wrong answer? lol

My understanding is that semivariance is the (partial) solution to returns that are asymmetric. It’ll still be a problem with fat tails and… cr*p… ok… semivariance isn’t so great to use with embedded options…

bchadwick Wrote: ------------------------------------------------------- > My understanding is that semivariance is the > (partial) solution to returns that are asymmetric. > It’ll still be a problem with fat tails and… > cr*p… ok… semivariance isn’t so great to use > with embedded options… so i think this implies chad is of opnion semi-variance is the right answer

semi-variance also could be, it can be applied to assymetric distributions but may not be the greatest meausre to use as far as contingent claip/cap risk question, that one I was sure off contingent claim risk - Yes cap risk - No

simple - semivariance is a poor measure when returns are skewed. all the other options were correct.

From Schwesser: Despite its advantages, Semi-variance is not a commonly used risk measure for the following reasons: - It is difficult to compute for large bond portfolio. While computing the portfolio std deviation is is computationally straightforward, there is no easy way of doing so for semi-variance …and after all that I am foggy about my answer, since it was a guess (by elimination).

volkovv Wrote: ------------------------------------------------------- > csk and hala, I am not standing to strongly by my > VAR answer, that seemed correct to me as well, > just some minutiae in the wording felt weird on > it. > > But as far as shortfall, I somewhat disagree with > you, it does give you the number but that number > is given with probability attached, so whever you > do exp. retun - 2 std (assuming normal > distribution), then you would say that the > expected shortfall would be that number with 95% > probability volkovv check definition of shortfall in investopedia - no probability attached

there is an end of chapter question in cfa or schweser where they clearly say that a mbs has cap risk don´t have the materials here, but i will check at home where. but, really, i am pretty sure of this because some weeks ago I didn´t understand why

fuck the investopedia… look at the CFAI material… and schweser tooo comp_sci_kid Wrote: ------------------------------------------------------- > volkovv Wrote: > -------------------------------------------------- > ----- > > csk and hala, I am not standing to strongly by > my > > VAR answer, that seemed correct to me as well, > > just some minutiae in the wording felt weird on > > it. > > > > But as far as shortfall, I somewhat disagree > with > > you, it does give you the number but that > number > > is given with probability attached, so whever > you > > do exp. retun - 2 std (assuming normal > > distribution), then you would say that the > > expected shortfall would be that number with > 95% > > probability > > volkovv check definition of shortfall in > investopedia - no probability attached

Ozzy05 Wrote: ------------------------------------------------------- > simple - semivariance is a poor measure when > returns are skewed. all the other options were > correct. Semivariance is a good measure when things are skewed, assuming that (like most investors), you’re more worried about downside risk. When things are skewed, you only care about the distribution of risk below the expected value, and semivariance captures this better than variance. It’s the fat tails that makes variance not so good to use. Options give you (very) fat tails.

lol, I remembered it too. There was interest rate risk, call risk and cap risk. I just didn’t remember what that chapter was about. hala_madrid Wrote: ------------------------------------------------------- > there is an end of chapter question in cfa or > schweser where they clearly say that a mbs has cap > risk > > don´t have the materials here, but i will check at > home where. but, really, i am pretty sure of this > because some weeks ago I didn´t understand why

hala_madrid Wrote: ------------------------------------------------------- > there is an end of chapter question in cfa or > schweser where they clearly say that a mbs has cap > risk > > don´t have the materials here, but i will check at > home where. but, really, i am pretty sure of this > because some weeks ago I didn´t understand why i remember the question on hand had a fixed rate? i think cap risk applies to floaters with a upside cap

Semi variance is, I believe, the correct answer because it is still impacted by some of the weaknesses of standard deviation (for a bond portfolio, duration changes over time and past returns are not an indicator of future returns, partly because credit events are low frequency).

It’s MBS, it can’t possibly have a fixed rate.

From Schwesser: Shortfall risk measures the probability that the actual return will be less than the target return. For example: the shortfall risk maybe be specified as: there is a 9.3% chance that returns will be less than T-bill rate this year. There are no massive disadvantages to calculating this measure

If that is true I can get ready to hang myself!

My recollection is that the statement the person made regarding shortfall risk wasn’t even the right definition? Am I thinking about a different question?

SEMI-VARIANCE all the way… and btw i know it coz i manage a bond portfolio…

It was the right definition in one way. But it looks like shortfall risk can be defined differently according to CSK and other guys. emarkhans Wrote: ------------------------------------------------------- > My recollection is that the statement the person > made regarding shortfall risk wasn’t even the > right definition? Am I thinking about a different > question?