preview background for page pf1
Mean-Variance Portfolio Analysis
Ilya Dergunov
Theory of Finance
Higher School of Economics
Module 3, 2023
preview background for page pf2
Outline
Outline
This Lecture
Mean-Variance Analysis: Nrisky assets
Mean-Variance Analysis: Nrisky assets + risk-free asset
Mean-Variance Portfolio Optimization
Ilya Dergunov 1
preview background for page pf3
MV: NRisky Assets
Mean-Variance Analysis: NRisky Assets
Mean-Variance Portfolio Optimization
Ilya Dergunov 2
preview background for page pf4
MV: NRisky Assets
The Mean-Variance Portfolio Problem
Suppose there are Nrisky assets available for investment
let µdenote the vector of asset expected returns and Σ the
variance-covariance matrix of returns.
Then, for any portfolio w, expected return and portfolio
variance are given by
µp=wµ σ2
p=wΣw
The minimum-variance portfolio with expected return µpis
the solution w(µp) to
min
w
1
2wΣw
s.t.1w= 1
µw=µp
Mean-Variance Portfolio Optimization
Ilya Dergunov 3
preview background for page pf5
MV: NRisky Assets
The Mean-Variance Portfolio Problem (cont.)
To solve this problem, set up the Lagrangian
L=1
2wΣw+λ1(µpµw) + λ2(1 1w)
The FOCs are
L
w= Σwλ1µλ21= 0
L
λ1
=µpµw= 0
L
λ2
= 1 1w= 0
Hence, all minimum-variance portfolios are of the form
w=λ1Σ1µ+λ2Σ11
Mean-Variance Portfolio Optimization
Ilya Dergunov 4
preview background for page pf6
MV: NRisky Assets
The Mean-Variance Portfolio Problem (cont.)
FOC already presents interesting result
Mean-variance efficient portfolio is constructed on the basis of
two mutual funds, based on Σ11and Σ1µ
The first mutual fund based on Σ11will have the intuition of
being the global minimum variance portfolio
The second fund focuses instead on obtaining the highest
expected return per unit of standard deviation
entire m-v frontier is spanned by these two intuitive mutual
funds
Mean-Variance Portfolio Optimization
Ilya Dergunov 5
preview background for page pf7
MV: NRisky Assets
The Mean-Variance Portfolio Problem (cont.)
In order to determine the constants λ1and λ2, just use the
two constraints and require that they be satisfied
For the first constraint, we have
µw=µ(λ2Σ11+λ1Σ1µ) = µp
or
preview background for page pf8
preview background for page pf9
preview background for page pfa
preview background for page pfb
preview background for page pfc
preview background for page pfd
preview background for page pfe
preview background for page pff
preview background for page pf10
preview background for page pf11
preview background for page pf12
preview background for page pf13
preview background for page pf14
preview background for page pf15
preview background for page pf16
preview background for page pf17
preview background for page pf18
preview background for page pf19
preview background for page pf1a
preview background for page pf1b
preview background for page pf1c
preview background for page pf1d
preview background for page pf1e
preview background for page pf1f