Hello @tattoo
Yes, this is a small typo error in the notes. The sentence should read, "it should monitor those limits to make sure they are followed"
Errors Found in Study Materials P1.T1. Foundations (OLD thread)
- Thread starter Nicole Seaman
- Start date
- Status
Yes, this is a small typo error in the notes. The sentence should read, "it should monitor those limits to make sure they are followed"
HI @tattoo Yes totally correct about both typos. Thank you! @Nicole Seaman I highlighted these fixes in red below (page 5):
FRMNinjaLeonardo
New Member
On page 19 of the study material of Chapter 5 (MPT & CAPM)
@David Harper CFA FRM can you confirm? This matches the Amenc source on page 111, as shown below:
(Noel Amenc and Veronique Le Sourd, Portfolio Theory and Performance Analysis (West Sussex, England: John Wiley & Sons, 2003))
Thank you @Nicole Seaman good find! To your point, I think the note is correct as it is @FRMNinjaLeonardo I think the confusion is that the last term (second bullet) contains σ(M) divisor, but this is the result of taking Jensen's alpha:
E(Rp) - Rf = α + β*[E(Rm) - Rf] where β = ρ*σ(p)/σ(m)
... but we assume ρ = 1 such that β is approximated by β ≈ σ(p)/σ(m) and Jensen's alpha:
E(Rp) - Rf = α + σ(p)/σ(m)*[E(Rm) - Rf] and then we can divide each side by σ(p):
(E(Rp) - Rf)/σ(p) = α/σ(p) + (σ(p)/σ(m)*[E(Rm) - Rf])/σ(p) but in the last term it cancels:
(E(Rp) - Rf)/σ(p) = α/σ(p) + (σ(p)/σ(m)*[E(Rm) - Rf])/σ(p) giving that final result:
(E(Rp) - Rf)/σ(p) = α/σ(p) + [E(Rm) - Rf])/σ(m)
Please note that we do not normally or necessarily assume ρ = 1. This is as assumption made by Amenc for purposes of this simplification where he invokes a certain definition of "well diversified." I would not want folks to think that "diversified" necessarily implies ρ(p, M) = 1.0, although we do have a definition of diversification as imperfect correlation.
E(Rp) - Rf = α + β*[E(Rm) - Rf] where β = ρ*σ(p)/σ(m)
... but we assume ρ = 1 such that β is approximated by β ≈ σ(p)/σ(m) and Jensen's alpha:
E(Rp) - Rf = α + σ(p)/σ(m)*[E(Rm) - Rf] and then we can divide each side by σ(p):
(E(Rp) - Rf)/σ(p) = α/σ(p) + (σ(p)/σ(m)*[E(Rm) - Rf])/σ(p) but in the last term it cancels:
(E(Rp) - Rf)/σ(p) = α/σ(p) + (
(E(Rp) - Rf)/σ(p) = α/σ(p) + [E(Rm) - Rf])/σ(m)
Please note that we do not normally or necessarily assume ρ = 1. This is as assumption made by Amenc for purposes of this simplification where he invokes a certain definition of "well diversified." I would not want folks to think that "diversified" necessarily implies ρ(p, M) = 1.0, although we do have a definition of diversification as imperfect correlation.
Last edited:
FRMNinjaLeonardo
New Member
@David Harper CFA FRM I got it! thanks.
A new thread has been created here for errors found in 2021 Topic 1 materials: https://forum.bionicturtle.com/threads/errors-found-in-2021-study-materials-p1-t1-foundations.23681/
- Status
Similar threads
- Sticky
- Sticky
- Sticky
- Sticky