Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 36 views

    Optimal Portfolios

    Uploaded by

    leolau2015
    The document discusses optimal portfolios and introduces the concept of a risk-free asset. It shows that combining the risk-free asset with a risky portfolio results in a capital allocation line (CAL) between them. The optimal risky portfolio is the one that results in the CAL with the highest slope, and all efficient portfolios lie along this CAL.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Optimal Portfolios

    Uploaded by

    leolau2015
    36 views32 pages
    The document discusses optimal portfolios and introduces the concept of a risk-free asset. It shows that combining the risk-free asset with a risky portfolio results in a capital allocation line (CAL) between them. The optimal risky portfolio is the one that results in the CAL with the highest slope, and all efficient portfolios lie along this CAL.

    Original Description:

    Optimal Portfolios

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses optimal portfolios and introduces the concept of a risk-free asset. It shows that combining the risk-free asset with a risky portfolio results in a capital allocation line (CAL) between them. The optimal risky portfolio is the one that results in the CAL with the highest slope, and all efficient portfolios lie along this CAL.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    36 views32 pages

    Optimal Portfolios

    Uploaded by

    leolau2015
    The document discusses optimal portfolios and introduces the concept of a risk-free asset. It shows that combining the risk-free asset with a risky portfolio results in a capital allocation line (CAL) between them. The optimal risky portfolio is the one that results in the CAL with the highest slope, and all efficient portfolios lie along this CAL.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 32

    Lecture 5

    Optimal portfolios

    Learning outcomes
    By the end of this lecture you should:
    Be familiar with the separation theorem
    Know why this implies that every investors
    optimal risky portfolio is the market portfolio
    Be able to solve the portfolio problem in the
    presence of a risk free asset and the market
    portfolio
    Understand how this implies the CAPM

    The optimal portfolio


    Last time we found the optimal portfolio by picking
    the portfolio on the efficient frontier that touched
    our highest indifference curve (gave the highest
    utility).
    E(r)

    Introduce a risk free asset


    Suppose that in addition to the risky assets that
    we talked about last lecture, we could also invest
    in some risk free asset
    Well call the return of this asset the risk free
    return, rf
    By definition of risk free, we have Var rf 0 E r r
    During our first three lecture we were only
    concerned with these kind of assets
    For simplicity we will assume that we have a flat
    and constant term structure of interest rates
    f

    Combining the risk free asset with a risky


    portfolio into a complete portfolio
    Suppose we want to combine some risky
    portfolio, P, with the risk free asset
    Lets denote the fraction invested in the risky
    asset y and the fraction invested in the risk
    free asset (1-y)
    The return of our complete portfolio, C, is
    defined as before:

    rC 1  y rf  yrP

    The expected return of a complete


    portfolio
    As usual, we are interested in the risk and
    expected return of this portfolio. Lets start with
    the expected return:
    E r E > 1  y r  yr @

    E r 1  y r  yE r

    Like before the portfolio expected return is a
    weighted average of the asset expected returns
    It will be convenient to express this equation as:
    E r r  y>E r  r @

    By varying y, we can choose our expected return
    C

    The risk of a complete portfolio


    We figured out in the last lecture how to calculate
    the variance of a two asset portfolio:
    r 1  y r  yr

    Var r Var > 1  y r  yr @

    Var r 1  y Var r  y Var r  2 1  y yCov r , r
    The neat thing here is that rf is a constant, so Var(rf) =
    0 and Cov(rf, rP) = 0:

    Var r y Var r y V
    V
    yV
    yV

    We find that C is linear in P
    C

    2
    P

    2
    P

    These are linear combinations


    We have found that:

    E rC rf  y>E rP  rf @
    V C yV P

    So y determines what fraction of the distance between
    the risk free asset and P is covered in both dimensions
    When y = 0 we are in the risk free asset, e.g. C = 0 and
    E(RC) = rf
    When y = 0.5 we are halfway between the risk free asset
    and P, e.g. C = 0.5P and E(RC) = rf + 0.5[E(RC) - rf]
    When y = 1 we are in portfolio P, e.g. C = P and E(RC) =
    E(RP)

    Lets plot it
    Graphically, this means that all our complete portfolios
    plot on a straight line between the risk free asset and P
    Lets call the line associated with the risky portfolio P
    for CALP (for reasons that will become clear later)
    E(r)
    P

    CALP

    rf

    Lets plot it
    Last lecture, we learned only to consider risky
    portfolios on the efficient frontier, so lets
    chose P from that set

    E(r)
    P

    CALP

    rf

    Apply the mean variance criterion


    We see that some of our complete portfolios
    dominate some portfolios on the efficient frontier
    Which ones are dominated depends on the risky
    portfolio we choose
    P2

    E(r)

    CALP2

    CALP1
    rf

    P1

    Apply the mean variance criterion


    Were interested in choosing the P that dominates the most portfolios
    This turns out to be the portfolio where the line of complete portfolios is
    tangent to the efficient frontier
    We call this P the optimal risky portfolio, P*
    The associated line CALP* is often simply denoted CAL

    CALP* = CAL
    E(r)

    P*

    rf

    Apply the mean variance criterion

    A different way to phrase this is to note that we only consider risky portfolios on the efficient
    frontier
    We can then forget about the efficient frontier and only compare CALs
    We note that for any portfolio on CAL1 there is a dominating portfolio just above it on CAL2
    Portfolios on CALs with higher slopes will always dominate portfolios on CALs with lower
    slopes
    The optimal risky portfolio is the portfolio associated with the CAL that has the highest slope

    CAL2
    E(r)
    CAL1
    rf

    The optimal risky portfolio


    All portfolios other than P* on the efficient
    frontier are dominated by some combination of
    the optimal risky portfolio, P*, and the risk free
    asset
    This means that all efficient portfolios consist of
    some such combination
    The reason we call the corresponding line CAL is
    that all capital will be allocated along it
    CAL is an acronym for the Capital Allocation Line

    The separation theorem


    All efficient portfolios (in the presence of a risk free
    asset) are on the CAL
    The CAL is determined by the optimal risky portfolio
    We pick the portfolio on the CAL that offers the
    amount of risk we want to take. This is expressed in our
    choice of y.
    Thus, our entire portfolio choice problem can be
    separated into two parts:
    Find the optimal risky portfolio, P*
    Choose how much risk we want by choosing the fraction of
    our wealth that we invest in that portfolio, y

    Step 1: Choosing P* (and implicitly the


    CAL)
    The CAL is a line in risk-return space
    Its slope, S, determines how much reward in terms of
    E(r) we get for taking on one more unit of risk
    We can easily calculate this slope from our two known
    points on the line:

    'y E r  E r E r  r
    S
    V 0
    V
    'x

    We usually refer to the slope of a CAL associated with a
    given portfolio as the Sharpe ratio of that portfolio


    P*

    P*

    P*

    P*

    Step 1: Choosing P*
    Recall from the construction of our utility function that
    we only care about E(r) and
    This means we always prefer a higher Sharpe ratio, S
    We find the optimal portfolio, P*, by choosing its
    portfolio weights, wP, so as to maximizes SP:

    E r  r
    max S
    V

    This gets messy (at least when we can choose between
    many assets)
    You can do this using the Excel solver or some other
    suitable computer program


    P

    wP

    Step 2: Choosing the risky share, y


    We choose y according to our risk preference, which is
    modeled in our utility function by A
    We may again illustrate this choice using indifference curves:


    E(r)

    P*
    C

    rf

    Step 2: Choosing the risky share, y


    Once we have determined P*, we know from
    before that our risk and return will be:
    >

    E rC rf  y E rP  rf

    VC

    yV P

    We also know that our utility will depend on


    these quantities in the following manner:
    U E r  1 2 AV 2

    Lets combine these equations:
    U r  y>E r  r @ A yV r  y>E r  r @ Ay V
    f

    P*

    P*

    P*

    2
    P*

    Step 2: Choosing the risky share, y


    By choosing y, we choose where to end up on the CAL
    Lets choose y so as to maximize our utility:
    max U rf  y>E rP*  rf @ 1 2 Ay 2V P2*

    y
    We set the first derivative equal to zero and solve for y:
    wU

    >E r  r @ AyV 0
    wy

    E r  r
    1 E r  r
    y

    V
    V
    A
    A

    E r  r
    is known as the reward-to-risk ratio (and has an interpretation that
    V
    is very similar to the Sharpe ratio)
    y* is increasing in this ratio, meaning that the more rewards in terms of
    E(r) we get for taking on extra risk, the more we invest in the risky
    portfolio
    y* is decreasing in A, meaning that the more risk-averse we are, the less
    we invest in the risky portfolio
    P*

    P*

    2
    P*

    P*

    2
    P*

    2
    P*

    P*

    2
    P*

    Leveraged positions
    There is nothing in principle that prevents us
    from choosing y > 1
    This means that well take a short position in
    the risk-free asset, i.e. (1 - y) < 0
    The interpretation of this is that we borrow
    money
    We say that we take a leveraged position in P*

    Borrowing constraints
    In practice, we must borrow at a higher rate
    than we can invest at
    This is because lending money to us is not
    really risk free
    Graphically we get a kink in the CAL when y =
    1
    Since wed have higher default risks for more
    leveraged positions, the CAL may also be
    concave when y > 1

    Implications of the separation theorem


    The rational way to increase risk taking is to
    increase leverage (not to buy more tech
    stocks)
    All investors will end up holding the same risky
    portfolio
    Since prices adjust to set the supply of stocks
    equal to the demand for stocks, the portfolio
    demanded must be the portfolio supplied
    P* is the market portfolio, M

    Implications of the separation theorem


    The attractiveness of a stock is determined by
    its risk and return effects on this portfolio
    We saw last lecture that the expected return
    effect of a stock on a portfolio is linear and
    that the risk effect depends crucially on its
    covariance with the other stocks in the
    portfolio

    A note on expected returns

    A company that issues a stock is basically selling a claim to its future


    profits
    These profits are determined by the operations of the company
    Since the profits are risky, the company has to sell the claims at a price
    that is lower than their expected value
    If the price is lower, the expected return of the investors is higher
    Since these high expected returns are used to induce investors to hold
    unattractive risky stocks, high expected returns signify unattractive
    stocks, which may seem counterintuitive
    Of course, high expected returns are not themselves unattractive
    In equilibrium, expected returns are set so as to make all stocks equally
    attractive
    Some times we emphasize this by referring to a stocks expected return as
    its required return (to make it as attractive as all other stocks)

    The market portfolio


    Recall that the market portfolio is the optimal
    risky portfolio for all investors
    Each investor buys a small fraction of the
    portfolio
    The entire market portfolio simply consists of all
    assets
    Just like in other portfolios, the weight of each
    asset in the market portfolio is the assets total
    market value divided by the total value of the
    portfolio:
    V
    wi

    V
    j

    The expected return of the market


    portfolio

    The return of the market portfolio, rM, is


    N

    wr
    We calculate expectations just like with any other portfolio
    N
    E rM wi E ri

    rM

    i i

    i 1

    i 1

    It is clear that the contribution of asset i to the expected return of the


    market portfolio is
    wi E ri

    It will be useful to express this as the contribution of asset i to the market
    portfolios excess return, i.e. its return over and above the risk-free rate
    N

    E rM  rf wi >E ri  rf @
    i 1

    The contribution of asset i to the market excess return is

    >

    wi E ri  rf

    The variance of the market portfolio


    We calculate the variance just like any other
    portfolio variance, i.e. by setting up the
    covariance matrix and summing the elements:



    Note that every asset corresponds to one row
    in the matrix

    w1r1

    w2r2

    wNrN

    w1r1

    Cov(w1r1,w1r1)

    Cov(w1r1,w2r2)

    Cov(w1r1,wNrN)

    w2r2

    wNrN

    Cov(w2r2,w1r1)

    Cov(wNrN,w1r1)

    Cov(w2r2,w2r2)

    Cov(wNrN,w2r2)

    Cov(w2r2,wNrN)

    Cov(wNrN,wNrN)

    The variance of the market portfolio


    The contribution of each asset to the variance
    of the market portfolio is captured by the sum
    of the elements in its row
    Lets view the row for asset i in isolation:

    This matrix corresponds to the matrix we
    would set up to calculate

    wiri

    w1r1

    w2r2

    wNrN

    Cov(w1r1,w1r1)

    Cov(w1r1,w2r2)

    Cov(w1r1,wNrN)

    Cov wi ri , w j rj
    j 1

    The risk-return ratio of the market


    portfolio
    We see that the contribution of asset i to the
    variance of the market portfolio is

    Cov w r , w r Cov w r , r w Cov r , r


    Note that the reward-to-risk ratio of the market


    portfolio is:
    E r  r
    V

    The contribution of asset i to this ratio is:
    N

    i i

    j j

    i i

    j 1

    2
    M

    >

    wi E ri  rf

    wi Cov ri , rM

    E ri  rf

    Cov ri , rM

    The risk-return ratio of the market


    portfolio
    Since the market portfolio is the portfolio with
    the best risk-return ratio, it cannot be
    improved by changing the portfolio weights
    This means that no isolated investment can
    make a larger contribution to the risk-return
    ratio than any other investment:
    E r  r
    E r  r

    Cov r , r Cov r , r
    This is also true for the market portfolio itself:
    i

    E ri  r f

    Cov ri , rM

    E rM  rf

    Cov rM , rM

    E rM  rf

    V M2

    The CAPM
    We can rewrite this equation as
    E r  r
    E r  r

    V
    Cov r , r
    Cov r , r

    E r  r
    E r r 
    V
    This equation expresses a relation that must hold
    between an assets expected return and its
    covariance with the market
    We call this model the Capital Asset Pricing
    Model or the CAPM
    It will be the focus of our coming lectures
    i

    2
    M

    i
    2
    M

    You might also like