I also believe borrower qualiy is the correct pick.
Etienne Wrote: > > It can’t be core-satellite since it said the guy > wanted to minimise misfit risk, which is a > core-satellite problem. > > I went with Port alpha, but I think it should have > been stock-based enhanced indexing. > > I was thinking maybe he can use sell an ETF/future > to get rid of the maket risk. Then he would still > only be long equities. This doesn’t fit with the > no derivs or no short constraints. Common sense > dictates stock-based enhanced. It actually was core-satelite for exactly the reason you described above. The guy wanted to minimize misfit risk, and that is exactly the property of core-staelite strategy that it does minimize misfit risk, hence it is exactly what the client wanted and the right answer. Usually minimizing misfit risk is considered a disadvantage of core-satelite approach, because misfit risk is a source of additional excess return for the manager, but that perspective was irrelevant for this particular question.
Borrower quality is correct answer. I chose it initialy but then double guessed myself and changed it to interest in the economy (applying that logic that since it is not fed funds rate, its to broad of a description). I checked the CFAI books and all three choices are sited there word for word as factors impacting repo rate, and borrower quality is not mentioned. Quality of Collateral Term of the Repo Interest in the Economy I agree with people who tried to make a case otherwise, Bear is perfect example. But we gotta admit for the exam we are living in CFAI world and playing by their rules, and borrower quality was the wrong answer for that reason. -1 for me
volkovv Wrote: ------------------------------------------------------- > Borrower quality is correct answer. I chose it > initialy but then double guessed myself and > changed it to interest in the economy (applying > that logic that since it is not fed funds rate, > its to broad of a description). > > I checked the CFAI books and all three choices are > sited there word for word as factors impacting > repo rate, and borrower quality is not mentioned. > > > Quality of Collateral > Term of the Repo > Interest in the Economy > > I agree with people who tried to make a case > otherwise, Bear is perfect example. But we gotta > admit for the exam we are living in CFAI world and > playing by their rules, and borrower quality was > the wrong answer for that reason. > > -1 for me -7 for me total. You are still doing pretty good 
volkovv Wrote: ------------------------------------------------------- > Borrower quality is correct answer. I chose it > initialy but then double guessed myself and > changed it to interest in the economy (applying > that logic that since it is not fed funds rate, > its to broad of a description). > > I checked the CFAI books and all three choices are > sited there word for word as factors impacting > repo rate, and borrower quality is not mentioned. > > > Quality of Collateral > Term of the Repo > Interest in the Economy > > I agree with people who tried to make a case > otherwise, Bear is perfect example. But we gotta > admit for the exam we are living in CFAI world and > playing by their rules, and borrower quality was > the wrong answer for that reason. > > -1 for me are you sure in the textbook it says interest in the economy and not fed funds rate?
Did anyone know the answer for managed futures?
mo34 Wrote: ------------------------------------------------------- > I don’t have the books around anymore, but if you > look at the end-of-chapter problems, you will find > a problem where they ask you to identify the best > way to execute a trade, look at the last one ( > done with a principal trade), check out their > reasons in the solutions, they all apply in this > case. > > Essentially the bottom line is if the trade is so > large ( like in this case), use someone to shop > it around for you. I did not remember reading that > they had 6 month to execute the trade, but if it’s > the case, then I am not sure what the solution > would be. I don’t like ECN because of the high > probability of having the trade unfilled, if most > of the company is owned by institutions ( Long > term investors, not traders), you’ll have a hard > time filling the trade on an ECN. Institutions are the parties ECNs are made for. That one had to be ECNs.
actuaryalfred Wrote: ------------------------------------------------------- > suddenly remember another question, like using a > coupon bond with the exact duration as the > liability or two bonds with +1 -1 year of > maturities and same duration, which one is best > according to classical immunization theory Wasn’t the answer on that one the one that had the closest duration (exactly even, maybe) AND was a zero?
volkovv Wrote: ------------------------------------------------------- > Borrower quality is correct answer. I chose it > initialy but then double guessed myself and > changed it to interest in the economy (applying > that logic that since it is not fed funds rate, > its to broad of a description). > > I checked the CFAI books and all three choices are > sited there word for word as factors impacting > repo rate, and borrower quality is not mentioned. > > > Quality of Collateral > Term of the Repo > Interest in the Economy > > I agree with people who tried to make a case > otherwise, Bear is perfect example. But we gotta > admit for the exam we are living in CFAI world and > playing by their rules, and borrower quality was > the wrong answer for that reason. > > -1 for me For what it’s worth, the first thing listed in Schweser is credit risk (not that it’s listed in priority, but it is the first thing mentioned as affecting the repo rate). I’m still pretty confident on quality of collateral - it matters, just matters the least.
former trader Wrote: ------------------------------------------------------- > volkovv Wrote: > -------------------------------------------------- > ----- > > Borrower quality is correct answer. I chose it > > initialy but then double guessed myself and > > changed it to interest in the economy (applying > > that logic that since it is not fed funds rate, > > its to broad of a description). > > > > I checked the CFAI books and all three choices > are > > sited there word for word as factors impacting > > repo rate, and borrower quality is not > mentioned. > > > > > > Quality of Collateral > > Term of the Repo > > Interest in the Economy > > > > I agree with people who tried to make a case > > otherwise, Bear is perfect example. But we > gotta > > admit for the exam we are living in CFAI world > and > > playing by their rules, and borrower quality > was > > the wrong answer for that reason. > > > > -1 for me > > > are you sure in the textbook it says interest in > the economy and not fed funds rate? It says the factor is “prevailing interest rates in the economy” and then the text speaks only of the federal funds rate which “is often said to represent prevailing interest rates in the United States on overnight loans.” The description doesn’t really match the text, or the idea. What if the repo is 30 days? Yeah, it will bear some relationship to the prevailing 30 day rate, but (a) it could have any number of spread relationships to the fed funds rate and (b) the answer was “general interest rates in the economy” which I took to mean a bunch of different interest rates of a bunch of different terms which could have a bunch of different relationships to each other. I don’t see how it can be read any other way.
thetank Wrote: ------------------------------------------------------- > emarkhans Wrote: > -------------------------------------------------- > ----- > > hezagenius Wrote: > > > -------------------------------------------------- > > > ----- > > > umuc Wrote: > > > > However if you actually calculate the > > quarterly > > > > returns and divide by 4, then you get the > > > second > > > > set of numbers. > > > > > > > > Now remember that GIPS post 2006 is monthly > > > > returns, not yearly. > > > > > > > > The question asked “closest” to. > > > > > > > > The monthly returns will be closer to the > > > > quarterly returns, hence the second set of > > > > numbers. > > > > J > > > > > > > > > Why would you divide by 4? If you do that, > you > > > are not geometrically linking the quarters. > I > > did > > > it by quarter and geometrically linked them > and > > > then just did year-end to year-end and got > the > > > same results because there were no CF in or > out. > > > > > I went with the first set of numbers. > > > > If you start a year with $1 and you end it with > > $1.12, your return for the year is 12% and it > > doesn’t matter how you get there. > > amen, no need to geometrically link anything, 2 > divisions and ur done I completely agree. You did not need to geometrically link anything because there were no intermediate CF’s. I just did it as a double check. I knew they would be the same. The question was one of the most straight-forward on the whole exam. I was quoting someone else who apparently did it differently and came up with another set of answers.
time spent on a question should be proportional to its straightforwardness
itstoohot Wrote: ------------------------------------------------------- > if the liabilities 5-7 year the best portfolio is > 2-10 year bonds or portfolio A?? The liabilities had a duration of 5.8. There was a portfolio that had a duration of 5.8 and bonds with 5-7 year maturities - wasn’t that the one to immunize the liabilities?
I also chose that one instead of the coupon bond with duration 5.8, forgot the other part of the question though
volkovv Wrote: ------------------------------------------------------- > Etienne Wrote: > > > > It can’t be core-satellite since it said the > guy > > wanted to minimise misfit risk, which is a > > core-satellite problem. > > > > I went with Port alpha, but I think it should > have > > been stock-based enhanced indexing. > > > > I was thinking maybe he can use sell an > ETF/future > > to get rid of the maket risk. Then he would > still > > only be long equities. This doesn’t fit with > the > > no derivs or no short constraints. Common > sense > > dictates stock-based enhanced. > > It actually was core-satelite for exactly the > reason you described above. The guy wanted to > minimize misfit risk, and that is exactly the > property of core-staelite strategy that it does > minimize misfit risk, hence it is exactly what the > client wanted and the right answer. > > Usually minimizing misfit risk is considered a > disadvantage of core-satelite approach, because > misfit risk is a source of additional excess > return for the manager, but that perspective was > irrelevant for this particular question. I thought about this, but if you are minimizing misfit risk against an ultra-largecap benchmark, then your satellite portfolios are going to be benchmarked to this. And yet, there’s almost no alpha in ultra-largecap, and therefore you’d be paying managers fees for almost no alpha. I didn’t think that sounded very palatable.
hezagenius Wrote: ------------------------------------------------------- > umuc Wrote: > > However if you actually calculate the quarterly > > returns and divide by 4, then you get the > second > > set of numbers. > > > > Now remember that GIPS post 2006 is monthly > > returns, not yearly. > > > > The question asked “closest” to. > > > > The monthly returns will be closer to the > > quarterly returns, hence the second set of > > numbers. > > J > > > Why would you divide by 4? If you do that, you > are not geometrically linking the quarters. I did > it by quarter and geometrically linked them and > then just did year-end to year-end and got the > same results because there were no CF in or out. > I went with the first set of numbers. Of course the question is whether GIPs wants you to geometrically link monthly returns to find annual return or take the average of monthly returns and multiply by 12? I have no idea about the answer about such fine points of GIPs.
All GIPS measures are annualized, except for periods less than one year, which are simple raw return from start to end. So geometric linking and just taking EV/BV-1 for December-to-December will give you the same answers.
bchadwick Wrote: ------------------------------------------------------- > All GIPS measures are annualized, except for > periods less than one year, which are simple raw > return from start to end. So geometric linking > and just taking EV/BV-1 for December-to-December > will give you the same answers. yeah it’s the same. i verified it during the exam.
bchadwick Wrote: ------------------------------------------------------- > All GIPS measures are annualized, except for > periods less than one year, which are simple raw > return from start to end. So geometric linking > and just taking EV/BV-1 for December-to-December > will give you the same answers. There are two ways of annualizing: 1) geometric linking 2) simply multiply by number of periods in a year The first method gave the first set of answers, the second method gave the second set of answers (I checked during the exam). You may think that the second method doesn’t make sense or is not used, but in reality it is widely used. Think bond yields. How do you report? You find the six-month yield and multiply by 2 to report annual yield. You don’t do geometric linking (that is (1+r)^2 -1) Think car financing rates. How is EAR computed? You take the real monthly rate and multiply by 12 instead of (1+r)^12 - 1. (This also gives the dealers a slight advantage as the returns from 1) are less than 2)) Trust CFAI to give a difficult question like this. We know data is to be reported monthly. To find annual rate do we multiply by 12 or do we geometrically link? J
umuc Wrote: ------------------------------------------------------- > bchadwick Wrote: > -------------------------------------------------- > ----- > > All GIPS measures are annualized, except for > > periods less than one year, which are simple > raw > > return from start to end. So geometric linking > > and just taking EV/BV-1 for > December-to-December > > will give you the same answers. > > There are two ways of annualizing: > > 1) geometric linking > 2) simply multiply by number of periods in a year > > The first method gave the first set of answers, > the second method gave the second set of answers > (I checked during the exam). > > You may think that the second method doesn’t make > sense or is not used, but in reality it is widely > used. > > Think bond yields. How do you report? You find the > six-month yield and multiply by 2 to report annual > yield. You don’t do geometric linking (that is > (1+r)^2 -1) > > Think car financing rates. How is EAR computed? > You take the real monthly rate and multiply by 12 > instead of (1+r)^12 - 1. (This also gives the > dealers a slight advantage as the returns from 1) > are less than 2)) > > Trust CFAI to give a difficult question like this. > We know data is to be reported monthly. To find > annual rate do we multiply by 12 or do we > geometrically link? > > J this is different, they tell you how much you have at the end of the period, it doesn’t matter what kind of calculation you do, if you started the year with 100 bucks and you ended up with 120 w/o any cashflows, you had a 20% gain, end of story.