Ann Oper Res

DOI 10.1007/s10479-017-2466-7

MULTIPLE OBJECTIVE OPTIMIZATION

Portfolio selection problem: a review of deterministic and

stochastic multiple objective programming models

Meryem Masmoudi1 · Fouad Ben Abdelaziz2

© Springer Science+Business Media New York 2017

Abstract The literature on portfolio selection mostly concentrates on computational anal-

ysis rather than on modelling efforts. In response, this paper provides a comprehensive
literature review of multiple objective deterministic and stochastic programming models for
the portfolio selection problem. First, we summarize different concepts related to portfolio
selection theory, including pricing models and portfolio risk measures. Second, we report the
mathematical models that are generally used to solve deterministic and stochastic multiple
objective programming problems. Finally, we present how these models can be used to solve
the portfolio selection problem.

Keywords Portfolio selection · Multiple objective programming · Multiple objective

stochastic programming

1 Introduction

Portfolio management deals with the selection of best portfolios in a context of volatile returns
due to random changes in future securities prices (Crundwell 2008). Therefore, investors are
always looking for securities that provide a good balance between return opportunities and
risk.
We start by reviewing basic definitions related to portfolio selection. We present the
capital asset pricing model (Sharpe 1964), quantifiers for portfolio risk and the well-known
Markowitz model (Markowitz 1952) for portfolio selection.

B Meryem Masmoudi
[email protected]
Fouad Ben Abdelaziz
[email protected]
1 College of Business Administration, University of Bahrain, P.O. Box 32038, Sakhir,
Kingdom of Bahrain
2 NEOMA Business School, Rouen Campus, Boulevard André Siegfried, 76130 Mont-Saint-Aignan,
France

123
 Ann Oper Res

1.1 The return

Return on investment is a measure that evaluates the worth of the investment. In a financial
market, the return on investment in security i at time t can be defined as the ratio of the net
gain to the cost of the investment, which can be expressed as follows (Rasmussen 2003):
pi,t − pi,t−1 + di,t
ri,t =
pi,t
where pi,t is the price of the security i at time t, pi,t − pi,t−1 is the capital gain over the period
[t − 1, t] for security i, and di,t is the dividend payment during the same period for security
i. Other formulas can be used to compute the return over time (Brentani 2004). For example,
some authors define returns as “excess returns,” which are the amount of investment return
above the risk free rate of return (Rasmussen 2003).
Let us denote by x = (x1 , ..., xn ) a vector whose component xi denotes the weight or
proportion of the investor’s wealth allocated to the ith security in the portfolio. Obviously,
n
we have xi = 1 and 0 ≤ xi ≤ 1, i = 1, ..., n. Any portfolio P can be characterized by a
i=1
vector x = (x1 , ..., xn ).
Mathematically, the total random return R P of the portfolio P can be written as follows:

n
RP = ri x i
i=1

where ri is the random return of the ith security in portfolio P at a future date.
To be able to estimate the return of portfolio P at a future date, we need to estimate the
securities’ future prices. These prices cannot be known in advance as they are random. There-
fore, several possible prices (or prices intervals) are defined, and a probability is assigned to
each of these prices. The portfolio return can be evaluated by its expected return:

n
E(R P ) = r̄i xi
i=1

where E(.) is the expected value of the random return and r̄i = E (ri ).
Calculating the security return seems to difficult, as we need to define a probability dis-
tribution for a set of plausible returns. In their Capital Asset Pricing Model (CAPM), Sharpe
(1964) and Lintner (1965) estimated the return using a simple linear regression model.

1.2 The capital asset pricing model

The main hypothesis in the CAPM is that investors focus only on the market portfolio return
R M when managing their portfolios, where the market portfolio consists of all securities and
the proportion invested in each security corresponds to its relative market value (Cohen and
Natoli 2003). The Sharpe model includes a set of simplifying assumptions that make the
CAPM applicable (Athanasoulis and Shiller 2000):
• The probability distribution of returns is homogenously anticipated by investors;
• There are no commissions, taxes or expenses for markets which are supposed perfect;
• Unlimited sums of money can be lent or borrowed by investors at the same interest rate,
which is equal to the risk-free rate;
• The mean variance criterion is maximized by risk-averse investors.

123
Ann Oper Res

The CAPM proposes that a security’s return can be described completely by a linear combi-
nation of a market return and its security’s co-variation (Rasmussen 2003):

ri = r f + βi (R M − r f ) (1)

where r f is the return of the risk free security and βi is known as Sharpe’s Beta. The Beta of
a security i is defined as the covariance between the security return and the market return:
Cov (ri , R M )
βi =
V ar (R M )
We may observe three possible absolute values of the beta (Lee et al. 2010):
• If the beta is equal to 1, then the security’s return moves with the market return,
• If the beta is less than 1, then the security’s return is less volatile than the market return,
• If the beta is greater than 1, then the security’s return is more volatile than the market
return.
From historical values of the stock, the value of Beta can be determined by re-arranging
Eq. (1) into the following form known as the market model (Crundwell 2008):

ri = α + βi R M

where the value of the coefficients α and βi are respectively the intercept and the slope of a
linear regression of the historical data for ri and R M .
The Beta can be used to measure the security risk, which is the risk that the return will
decrease due to moves in the market. In the following, we report some other measures of the
volatilities of returns.

1.3 Risk measures

Portfolio selection involves several forms of risk, including (Zenios and Ziemba 2006):
• Credit risk: risk of a nonpayment;
• Liquidity risk: risk related to non-availability of cash to support the investment activities;
• Operational risk: risk of losses due to operational errors, and;
• Business risk: risks due to volatility of security volumes.
For investors, an appropriate measurement of risk should quantify the chances that the actual
return of an investment will not be as expected (Huang 2008).
For example, the mean absolute deviation (MAD) measures the average deviation in abso-
lute terms around the mean of the distribution as follows (Rachev and Stoyanov 2008):

M AD P = E (|R P − E (R P )|)

Another dispersion measure that penalizes symmetrically both negative and positive devia-
tions from the mean is variance. Variance (σ P ) is the widely-used measure to quantify the
risk of a portfolio P as follows:
⎡ ⎤

  ⎢⎢
n 
n ⎥
⎥
σ P2 = E (R P − E(R P ))2 = ⎢ xi2 σi2 + xi x j σi j ⎥ = x t V x
⎣ ⎦
i=1 j=1
i = j

123
 Ann Oper Res

  
where σi j = E (ri − r̄i ) r j − r̄ j is the covariance of security i and security j and V is
the n × n covariance matrix:
⎛ 2 ⎞
σ1 σ12 · · · σ1n
⎜ σ21 σ · · · σ2n ⎟
2
⎜ 2 ⎟
V =⎜ . . . . ⎟
⎝ .. .. . . .. ⎠
σn1 σn2 · · · σn2
The higher are the variance and dispersion of the returns, the more uncertain the future returns
and therefore the riskier the portfolio.
Markowitz (1959) proposed semi-variance, which considers observations below the mean
as a measure of risk (Kandasamy 2008):
 
SV P = E (min {R P − E (R P ) , 0})2

While semi-variance is well suited to describe security risk, the most often used risk measure
is Morgan’s value at risk (VaR) (Jorion 2001). The VaR represents the predicted maximum
loss with a specified probability level ε over a certain period (Giannopoulos et al. 2005):

P (R P ≥ −V a Rε (R P )) = ε

If we take the case of a portfolio which is held with a 1 day 99% VaR equal to $1 million,
this means that over 1 day, there is a 1% probability that the portfolio will lose more than $1
million.
Despite its popularity, the VaR has the disadvantage of not being a sub-additive measure
and does not take account of severe losses beyond the VaR value (Filho 2006).
Risk can be influenced by different factors, such as income uncertainty, interest rates,
inflation, exchange rates, tax rates, etc. Markowitz (1952) observed that investors should
diversify their investment to reduce risk, and that an efficient portfolio is one where no added
diversification can lower the portfolio’s risk for a given expected return.

1.4 The Markowitz model

Markowitz’s (1952) portfolio selection model aims to select the least risky portfolio for a
given level of return:

Min σ P2 = x t V x

n
s.t. r̄i xi ≥ R
i=1
n
xi = 1
i=1
0 ≤ xi ≤ 1, i = 1, ..., n (2)

where R is the desired level of expected return.

The Markowitz model (2) is a quadratic program with a unique solution as the covariance
matrix V is a positive definite matrix (variances of risky portfolios are strictly positive).
Many studies have followed Markowitz’ analysis and used or extended mean variance
analysis for portfolio selection (see Liu 2004). The mean variance model (2) can be rewritten
differently to maximize the return for a given level of risk σ :

123
Ann Oper Res


n
Max r̄i xi
i=1
s.t. x V x ≤ σ 2
t


n
xi = 1
i=1
0 ≤ xi ≤ 1, i = 1, ..., n (3)
For a given level of return R, program (1) provides an efficient portfolio.
A portfolio that solves program (2) or (3) is called an efficient portfolio if it provides
minimum risk for a given return, or a maximum return for a given risk. The concept of efficient
optimal solution results from multiple objective programming and stochastic dominance
fields.

2 Multiple objective programming

A multiple objective program can be written in the following form (Ehrgott 2005):
 
Min f (x) = f 1 (x) , f 2 (x) , ..., f p (x)
s.t. x ∈ X ⊂ I R n (4)
where X is the set of feasible solutions and f 1 (x) , ..., f p (x) are p functions that evaluate
the performance of the decision variable x.
For p =1, the problem (4) is a uni-objective mathematical program. With a single objec-
tive function and under some convexity assumptions, the problem (4) has a unique optimal
solution. In the case of multiple objectives p > 1, problem (4) may not have a feasible solu-
tion that optimizes all objective functions at the same time because of the usually observed
conflict between objective functions (Evans 1984). The concept of Pareto efficient solution
is generally proposed for such situations.

2.1 Efficiency and non-dominance

The concept of efficiency can be defined as follows:

Definition (Chankong and Haimes): A feasible solution x ∗ ∈ X is called efficient or Pareto
optimal, if there is no other x ∈ X such that f (x) ≤ f (x ∗ ). If x ∗ is efficient, f (x ∗ ) is called
a non-dominated point. If x1 , x2 ∈ X and f (x1 ) ≤ f (x2 ), we say that x1 dominates x2 and
f (x1 ) dominates f (x2 ). The set of all efficient solutions x ∗ ∈ X is denoted X E and called
the efficient set. The set of all non-dominated points y ∗ = f (x ∗ ), where x ∗ ∈ X E , is denoted
Y N and called the non-dominated set.
The literature gives some mathematical properties of the efficient set X E and the non-
dominated set Y N (for more details see Chankong and Haimes 1983; Figueira et al. 2005).
These sets may be empty or defined via isolated points.
The literature also defines other forms of efficiency, such as weak efficiency, strict effi-
ciency and proper efficiency.
Definition (Chankong and Haimes): A feasible solution x ∗ ∈ X is called weakly efficient
(weakly Pareto optimal) if there is no x ∈ X such that f (x) < f (x ∗ ), i.e. f k (x) < f k (x ∗ )
for all k = 1, ..., p . The point y ∗ = f (x ∗ ) is then called weakly non-dominated.

123
 Ann Oper Res

A feasible solution x ∗ ∈ X is called strictly efficient (strictly Pareto optimal) if there is no

x ∈ X , x  = x ∗ such that f (x) < f (x ∗ ) . The weakly (strictly) efficient and non-dominated
sets are denoted, X wE (X s E ) and YwE , respectively
A feasible solution x ∗ ∈ X is called properly efficient, if it is efficient and if there is a real
number M > 0 such that for all i and x ∈ X satisfying f i (x) < f i (x ∗ ) there exists an index
j such that f j (x ∗ ) < f j (x) such that
f i (x ∗ ) − f i (x)
≤M
f j (x) − f j (x ∗ )
The corresponding point y ∗ = f (x ∗ ) is called properly non-dominated.

We note that Y N ⊂ YwN and that X s E ⊂ X E ⊂ X wE (Figueira et al. 2005).

It is difficult to explicitly generate all these sets correctly. The decision maker is usually
interested in one or few solutions. In the next subsections, we discuss solution strategies
used to solve the multiple objective program (4) to obtain a reasonable sample of efficient
solutions.

2.2 The weighted sum method

The most commonly used approach to solve the multiple objective program (4) is the weighted
sum method:

p
Min wi f i (x)
i=1
s.t. x ∈ X (5)
where wi is the weight for the objective function f i .
The problem (5) is easy to solve and it has been demonstrated that it can be used to generate
efficient solutions:

Proposition (Chankong and Haimes): Suppose that x ∗ is an optimal solution of the weighted
sum optimization problem (5) then one of the following statements hold:
1. If w ≥ 0 then x ∗ ∈ X wE .
2. If w > 0 then x ∗ ∈ X E .
3. If w ≥ 0 and x ∗ is a unique optimal solution of (5) then x ∗ ∈ X s E .
p
4. If w > 0 and wi = 1 then x ∗ is a properly efficient solution.
i=1

Let X be a convex set and f k , k = 1, ..., p; be convex functions. Then the following
statements hold:
1. If x ∗ is a properly efficient solution of (4) then there is some w > 0 such that x ∗ is an
optimal solution of (5).
2. If x ∗ is a weak efficient solution of (4) then there is some w ≥ 0 such that x ∗ is an optimal
solution of (5).

2.3 The ε -constraint model

The ε-constraint method consists in transforming p-1 objective functions into constraints by
fixing a threshold to the transformed functions as follows:

123
Ann Oper Res

Min f j (x)
s.t. f k (x) ≥ εk , k = 1, ..., p and k  = j
x∈X (6)
where εk is the threshold for the objective function f k , k = 1, ..., p and k  = j. We note
that a solution x ∗ is efficient if, and only if, there exist ε ∈  p such that this solution x ∗ is
optimal for the problem (6) for all j = 1, ..., p.
While it is computationally difficult to obtain an efficient solution for p programs (6) for
a given threshold ε ∈  p , in the following proposition we report results for other forms of
efficiency.

Proposition (Chankong and Haimes): Let x ∗ be an optimal solution of (6) for some j,
j = 1, ..., p, then x ∗ is weakly efficient. If x ∗ is the unique optimal solution of (6) for some
j, j = 1, ..., p, then x ∗ is a strictly efficient (and therefore efficient).

2.4 The goal programming model

The goal-programming model is similar to the ε-constraint method as objective functions

are transformed into constraints by setting up target values for these functions, called goals.
The goal programming model can be formulated as follows (Jones and Tamiz 2002):

p
Min α j d+ −
j + βj dj
j=1

s.t. f j (x) − d + −
j + d j = g j , j = 1, ..., p
x∈X (7)

where g j is the goal for the objective function f j ; d + −

j and d j are deviation variables rep-
resenting the under and over achievement of the jth goal respectively and α j and β j are
weights for positive and negative deviations of the jth objective functions f j , j = 1, ..., p.
Goal programming was introduced by Charnes et al. (1955) in an application of a single-
objective linear programming problem to estimate executive compensation. Nowadays, the
goal programming technique is commonly used for a long list of applications (Romero 1991).
The above program (7) is called the weighted goal programming model. Other goal pro-
gramming models exist, either related to the type of goal criteria (less than or equal to, greater
than or equal to, equal to, within a range) or to the aggregating approach between deviations
(weighted, lexicographic, minimax, etc.). For more details, see Jones and Tamiz (2002). In
all variants of the goal programming approach, the principle is to minimize deviations from
a target value. One of the goal programming variants is the compromise model.

2.5 The compromise programming model

The compromise program aims to minimize the deviation from the target values, which are
the ideal or the nadir point.

Definition (Ehrgott 2005): The point y I = (y1I , ..., y pI ) given by ykI = Min f k (x) is called
x∈X
the ideal of the multiple objective optimization problem (4). The point y N = (y1N , ..., y pN )
given by ykN = Max f k (x) is called the nadir of the multiple objective optimization prob-
x∈X
lem (4).

123
 Ann Oper Res

The compromise programming approach consists in finding the nearest solution to the
Nadir as follows (Aouni et al. 2005):

Min d( f (x), y N )
s.t. x ∈ X (8)

where d(., .) is a distance in I R p . When we choose the Euclidian L p distance, program (4)
is equivalent to the following program i (Ehrgott 2005):
  1p

p  p
Min wi f i (x) − yiN
i=1
s.t. x ∈ X

where wi is the weight for the ith deviation.

3 The multiple objective portfolio selection models

The Markowitz model (2) can be viewed as an epsilon constrained transformation of a bi-
objective model where the investor maximizes the expected return value (the first objective)
and minimizes the return variance (the second objective). These two objective functions
represent the first and second moment of the random return. In some situations, the investor
may be concerned with higher moments, for example skewness (Prakash et al. 2003).
Anagnostopoulos and Mamanis (2010) presented a portfolio selection model with three
objective functions: (i) to minimize the risk, (ii) to maximize the return and (iii) to minimize
the number of securities included in the portfolio. Xidonas et al. (2009) selected a portfolio
from the Athens stock exchange taking into account the following criteria: profitability (return
on security and return on equity), management performance (Asset turnover and Inventory
turnover) and capital structure (assets to liabilities and liabilities to equity). Ehrgott et al.
(2004) presented a multiple objective portfolio selection model where they replaced the risk
and return objectives by five objective functions: 12-month performance, 3-year performance,
the annual dividend or revenue, the Standard and Poor’s star ranking, and 12-month volatility.
Ida (2003) presented a multiple objective problem for the portfolio selection problem in which
some of the model coefficients are defined by intervals. The main idea was to consider that
the security return and variance are defined on an interval. The obtained model is a mean
variance model with an interval coefficient:

n
 
Max r inf i , r supi xi
j=1

n 
n
 
σ inf i j , σ supi j xi x j
i=1 j=1

n
s.t. xi = 1
i=1
xi ≥ 0, i = 1, ..., n

123
Ann Oper Res

 
where the return ri and the covariance σi j are restricted by intervals r inf i , r supi and
σ inf i j , σ supi j respectively.
Steuer et al. (2005) reported the following set of objective functions that were considered
in the literature for portfolio selection:
max {z 1 = portfolio return}
max {z 2 = dividends}
max {z 3 = amount invested in R&D}
max {z 4 = social responsibility}
max {z 5 = liquidity}
max {z 6 = portfolio return over that of a benchmark}
max {z 7 = −deviation from asset allocation percentage}
max {z 8 = −number of securities in portfolio}
max {z 9 = −turnover (i.e., costs of adjustment)}
max {z 10 = − max imum investment proportion weight}
max {z 11 = −amount of short selling}
max {z 12 = −number of sec urities sold short}
Some of the objective functions listed above are stochastic, and we discuss these stochastic
objective functions in a following section.

3.1 The weighted sum model

Xia et al. (2000) presented a model to improve the performance of the mean variance model
by considering the expected return of securities as decision variables:

n 
n 
n
Max (1 − w) r̄i xi − w σi j xi x j
i=1 i=1 j=1

n
s.t. xi = 1
i=1
r̄i ≥ r̄i+1 , i = 1, ..., n − 1
ai ≤ ri ≤ bi , i = 1, ..., n
xi ≥ 0, i = 1, ..., n (9)
where [ai , bi ] is the range in which the expected return of security i can vary. They solved pro-
gram (9) using genetic algorithms and reported a comparison with the traditional Markowitz
model to show the importance of the above model.

3.2 The goal programming model

Usually, objective functions in the portfolio selection problem (return, liquidity, risk, etc.)
have a target value or goal that we need to achieve (Aouni and Tarre 2010). This leads to
the construction of goal programming models for the portfolio selection problem (Azmi and
Tamiz 2010)
Below, we report a case study from the Tunisian security market, in which Mansour et al.
(2007) applied an interval goal programming model that considered three objective functions:

123
 Ann Oper Res

return, risk, in terms of the portfolio Beta and exchange flow ratio. For each of these objective
functions lower and upper target values are given (goals intervals) and a satisfaction function
is used to measure the difference between the actual value of the objective function and the
goal. The proposed model was written as follows:

3 
    
Max Z = w +j F j+ δ + − −
j + w j Fj δ−
j
j=1

n
s.t. r̄i xi − δ1+ + δ1− = ζ1
i=1

n
βi xi − δ2+ + δ2− = ζ2
i=1

n
L i xi − δ3+ + δ3− = ζ3
i=1

n
xi = 1
i=1
0 ≤ xi ≤ 0.1
 
ζ j ∈ glj , g uj ( f or j = 1, 2, 3)
δ+ −
j , δ j ≥ 0 ( f or j = 1, 2, 3) (10)

where F j+ (.) and F j− (.) are the satisfaction functions associated with respectively positive
δ+ − + −
j and negative deviations δ j ; wi and wi are weights associated respectively with positive
and negative deviations and g j , g j are respectively the lower and upper values for goal ζ j .
l u

3.3 The compromise programming model

Ballestero and Romero (1996) presented one of the first applications of compromise program-
ming model for portfolio selection. They proposed a compromise mean variance portfolio
selection model based on L p distance and defined as follows:

 ⎡ p  p⎤ 1
   n n  p
p

n
    
Min L p = ⎣w1  R − r̄i xi  + w2  σi j xi x j − V  ⎦
p
   
i=1 i=1 j=1

n
s.t. xi = 1
i=1
0 ≤ xi ≤ 1, i = 1, 2, ..., n (11)

where R and V are the ideal values for expected return and variance, respectively. Ballestero
and Romero (1996) studied the case where p = 1 and p = ∞. Ballestero and PlaSantamaria
(2003) applied the compromise model (11) to a set of securities from the Madrid security
market.

123
Ann Oper Res

3.4 Perspective and limitations of multiple objective portfolio selection models

According to Aouni et al. (2014), goal programming models are popular among researchers
for the portfolio selection problem because of the flexibility they offer concerning decision
maker preferences in terms of goals and importance of the criteria. Xidonas et al. (2012)
noted that these mathematical models are inefficient unless supported by algorithmic tools
and mechanisms. The decision maker is looking for a solution that can consider his/her
preferences and that takes account of the risk related to the return on the security, the nature
of the security and the state of the market (Al-Shammari and Masri 2015).

4 The multiple objective stochastic portfolio selection models

The return is primarily defined as a stochastic parameter, and the Markowitz model optimizes
the first and the second moment of this random return. In addition, almost all objective
functions reported by Steuer et al. (2005) are stochastic. In this section, we review stochastic
programming approaches and we review some of the single and multiple objective stochastic
models for portfolio selection.

4.1 Single and multiple objective stochastic programs

A stochastic program is a mathematical program where the objective function and/or the
constraints parameters are random variables (Kall and Wallace 1995):
Min f (x, w)
s.t. g j (x, w) ≥ 0, j = 1, ..., m
x∈X (12)
where f (x, w), g j (x, w) ≥ 0, j = 1, ..., m are functions defined using the random
parameter w, X is the set of feasible solutions defined by deterministic constraints and w is
a random event whose probability distribution P is defined on a set  ⊂ Rd .
A stochastic linear program can model problems under risk where explicit knowledge
of the probability distribution is the main hypothesis. An intuitive approach to solve the
stochastic program (12) is to replace the random parameters by their expected value and then
obtain the following equivalent mathematical program:
Min E ( f (x, w))
 
s.t. E g j (x, w) ≥ 0, j = 1, ..., m
x∈X
This approach is easy to implement but it may lead to an unfeasible decision (Sen and Higle
1999).
To solve the stochastic program (12), we need to transform it into an equivalent mathe-
matical program. A first strategy, called the wait and see approach, is to solve program (12)
for each realization of the random variable w and deliver the final solution after we observe
the value of the random parameter w (Bereanu and Peeters 1970). A second strategy, the
most used in practice and known as the here and now approach, supposes that we cannot wait
for the occurrence of the random event and we have to find a solution here and now. Of the
different here and now approaches, we describe in more detail below the chance constrained
approach and the recourse approach.

123
 Ann Oper Res

The chance constrained approach was first proposed by Charnes and Cooper (1963) where
a solution does not have to satisfy random constraints for all scenarios, but only for a percent-
age α of these scenarios. In a chance constrained approach, the equivalent of the stochastic
program (12) can be written as follows:

Min E ( f (x, w))

 
s.t. P g j (x, w) ≥ 0 ≥ α, j = 1, ..., m
x∈X (13)

where α ∈ [0, 1] is the probability (or reliability) level that characterizes the minimum degree
of satisfaction of the uncertain constraints and P [.] denotes the probability distribution.
Program (13) is called the joint chance constrained program because the probability level
α is the same for all random constraints. If the probability level is different from one constraint
to another, the resulting program, called the joint chance constrained program and is defined
as follows:

Min E ( f (x, w))

 
s.t. P g j (x, w) ≥ 0 ≥ α j , j = 1, ..., m
x∈X

where α j ∈ [0, 1] is the probability level related to the jth random constraints ( j =1,...,m).
Under some predefined condition for probability distribution, the chance constrained pro-
gram (13) has a convex set of feasible solutions (Prékopa 1995). In the case of normal
probability distribution, the set of feasible solutions in program (13) can be rewritten using
linear constraints (Ben Abdelaziz et al. 2007).
The recourse approach considers that all solutions in X are feasible solutions and if a
solution x ∈ X does not verify some of the uncertain constraints g j (x, w) ≥ 0, then we
penalize such a solution by introducing into the objective function an additional cost, called
the recourse function Q(x) (Birge and Louveaux 1997). The resulting certainty equivalent
program under a recourse approach for the stochastic program (12) can be written as follows
(Kall and Wallace 1995):

Min E ( f (x, w)) + Q(x)

s.t. x ∈ X (14)

where Q(x) = E (Q(x, w)) and

Q(x, w) = Min q(w)t y

s.t W (w)y = g − (x, w)
y≥0

where q(w) is the recourse cost, W (w) is the recourse matrix, y is the recourse decision and
g − (x, w) = g(x, w) if g(x, w) < 0 otherwise 0.
A multiple objective stochastic program is a stochastic program with more than two
objective functions. To solve a multiple objective stochastic program, we can follow one of
the following strategies (Ben Abdelaziz 2012):
• Transform the multiple objective stochastic program using one of the stochastic program-
ming approaches and then solve the resulting multiple objective program using one of
the multiple objective programming approaches;

123
Ann Oper Res

• Transform the multiple objective stochastic program using one of the multiple objective
programming approaches and then solve the resulting stochastic program using one of
the stochastic programming approaches.

Both strategies combine two kinds of transformations, a stochastic transformation and a

multiple objective transformation, to propose an equivalent mathematical program to the
multiple objective stochastic program.

4.2 Multiple objective stochastic portfolio selection models

Few scholars have discussed multiple objective stochastic models for portfolio selection.
In this section, we review some single and multiple objective stochastic portfolio selection
models.
Early on, Roy (1952) proposed the chance constrained approach to minimize the proba-
bility of the portfolio return being less than a predetermined disaster level R∗ :
 n 

Min P r i x i ≤ R∗
i=1

Xu et al. (2011) proposed to maximize the level of return R, such that the probability that
the portfolio return exceeds R is equal to α

Max R
 n 

s.t. Pr ri x i ≥ R ≥ α (15)
i=1

This is a chance constrained program (15), where the portfolio risk is considered through the
chance constraint.
Ben Tal (1991) proposed a two-stage recourse approach for the single objective stochastic
portfolio selection problem, where we first decide on the amount of money to invest in the
non-risky security x0 , and then invest the remaining wealth in risky securities. The model
was presented as follows:
 
n
Max E r0 x0 + ri x i
i=1

n
s.t x0 + xi = 1,
i=1
x0 ≥ 0, xi ≥ 0, i = 1, ..., n

Ben Abdelaziz et al. (2007) presented a chance constrained compromise programming

approach (CCCP) for the following bi-objective portfolio selection problem:


n
Max ri x i
i=1

n
O pt βi xi
i=1

123
 Ann Oper Res


n
s.t. xi = 1
i=1
0 ≤ xi ≤ u i , i = 1, ..., n (16)

where u i is the upper bound for the proportion to be invested in the security i, i = 1, ..., n.
The “Opt” in program (16) refers to the systematic portfolio risk that should be equal to a
predefined value. Ben Abdelaziz et al. (2007) assumed that their optimal portfolio should be
neither more nor less risky than the market, and therefore the target value for the portfolio
beta should be equal to 1.
The CCCP approach is a mix between the compromise programming approach (multiple
objective transformation) and the chance constrained programming approach (stochastic
transformation). The model was used to select a portfolio among 45 securities listed in
the Tunisian stock exchange market using the following three objective functions: rate of
return, exchange flow ratio (E Fi ) and portfolio beta. The resulting deterministic equivalent
mathematical program is written as follows:

Min Z = ε + δ1− + δ2− + δ3− + δ3+

 

45 
45
∗
s.t. R − r̄i xi + φ −1 (1 − ζ )σ ∗
R − ri x i − ε + δ1− = 0
i=1 i=1

45
E Fi xi + δ2− = E F ∗
i=1

45
βi xi + δ3− − δ3+ = 1
i=1

45
xi = 1
i=1
0 ≤ xi ≤ 0.1, i = 1, ..., 45
δ1− , δ2− , δ3− , δ3+ , ε ≥ 0

where R ∗ and E F ∗ are the ideal values of the rate of return objective function and the
exchange flow ratio objective function, respectively and φ is the probability distribution
function of the standard normal distribution.
Masmoudi and Ben Abdelaziz (2012) presented a recourse goal programming approach to
solve program (16), where the difference between the portfolio and the minimum acceptable
rate of return generates a penalty in the objective function. The resulting certainty equivalent
model for program (16) is as follows:


S
Min δ + + δ − + ps q (ws ) y (ws )
s=1

n
s.t. ri (ws ) xi + y (ws ) = R (ws ) , s = 1, ..., S
i=1

n
βi xi − δ + + δ − = 1
i=1

123
Ann Oper Res


n
xi = 1
i=1
0 ≤ xi ≤ u i , i = 1, ..., n
δ + ≥ 0, δ − ≥ 0, y (ws ) ≥ 0, s = 1, ..., S (17)
where the security random return ri (ws ) is defined over the discrete set of states of the market
 = {w1 , ...., w S } and ps is the probability of occurrence of the event ws and q (ws ) is the
recourse penalty for the event ws .
Recently, Masmoudi and Ben Abdelaziz (2015) extended their recourse goal programming
approach by penalizing the infeasible solution for uncertain constraints with the most probable
highest recourse cost rather than with the expected recourse cost. The proposed approach
is called the chance constrained recourse (CCR) approach, and the certainty equivalent to
program (16) is as follows:
Min δ − + δ + + ε

n
ε 
n
s.t. R − r̄i xi − + φ −1 (α) σi xi ≤ 0
q
i=1 i=1

n
βi xi + δ − − δ + = 1, i = 1, 2, ..., n
i=1

n
xi = 1
i=1
0 ≤ xi ≤ u i , i = 1, 2, ..., n
δ − ≥ 0, δ + ≥ 0, ε ≥ 0
Masri (2015) dealt with program (16) for cases where multiple stochastic goals are given for
the return objective function. The author proposed a chance constrained approach to address
the investors’ minimum acceptable rate of return and a recourse approach for the investors’
ideal rate of return.
The literature proposes few other models to build a certainty equivalent to the multiple
objective stochastic portfolio selection problem. Ben Abdelaziz et al. (2009) used the stochas-
tic goal programming approach to select a portfolio in the United Arab Emirates equity market
considering the following five objective functions: capital appreciation; current income; the
price earnings ratio; the market to book value ratio; and risk.
Masri et al. (2010) extended the CCCP approach for the following bi-objective stochastic
portfolio selection problem:

n
Max ri x i
i=1

n
Max li x i
i=1

s.t. xi ≤ Hr
i∈Sh

xi ≥ L r
i∈Sl

123
 Ann Oper Res


n
xi = 1
i=1
0 ≤ xi ≤ u i , i = 1, 2, ..., n

where li is the liquidity of security i, Hr and L r are the percentage to be invested in the set
Sh of high risk securities and the set Sh of low risk securities, respectively.

4.3 Perspective and limitations of multiple objective stochastic portfolio selection

models

The above multiple objective stochastic portfolio selection models may offer investors a
powerful tool for managing risky portfolios. Knowledge of the probability distribution was
an important assumption of the models presented. Such information does not hold in many
situations. Ben Abdelaziz and Masri (2010) proposed a compromise approach for a multiple
objective stochastic linear program in which the probability distribution is described by
partial linear equations. Investors are aware of the importance of risk and uncertainty in
today’s rapidly changing environments; they want to avoid making investments based on
models that disregard their ignorance of risk assessment.
Future research on the subject should explore the group decision aspect of multiple objec-
tive portfolio optimization. It should also address other forms of uncertainty while considering
the dynamic aspect of the investment process.

5 Conclusion

The Portfolio selection problem can be intuitively modeled as a multiple objective stochastic
program. This paper reviews the models currently proposed on the subject, and compares the
different assumptions and proposed solutions.
Today, the risk of not achieving a desired or planned portfolio return is the most challeng-
ing issue in portfolio selection. Researchers should develop models that reassess the risk in
a way that takes into account the sensitivity of the environment and the inherent ignorance
of market trends. New trends in portfolio selection models should consider uncertain proba-
bility distribution (Masri and Abdelaziz 2010), uncertain risk measures (Ben Abdelaziz and
Masmoudi 2014), socially responsible investment (Al-Shammari and Masri 2016), strategic
behaviors, group strategic behaviors and dynamic aspects.

References
Abdelaziz, F. B. (2012). Solution approaches for the multiobjective stochastic programming. European Journal
of Operational Research, 216(1), 1–16.
Abdelaziz, F. B., Aouni, B., & El Fayedh, R. (2007). Multi-objective stochastic programming for portfolio
selection. European Journal of Operational Research, 177, 1811–1821.
Abdelaziz, F. B., El Fayedh, R., & Rao, A. (2009). A discrete stochastic goal program for portfolio selection:
The case of United Arab Emirates equity market. INFOR, 47(1), 1–5.
Abdelaziz, F. B., & Masmoudi, M. (2014). A multiple objective stochastic portfolio selection problem with
random Beta. International Transactions in Operational Research, 21(6), 919–933.
Abdelaziz, F. B., & Masri, H. (2010). A compromise solution for the multiobjective stochastic linear program-
ming under partial uncertainty. European Journal of Operational Research, 202(1), 55–59.
Al-Shammari, M., & Masri, H. (2015). Multiple criteria decision making in finance, insurance and investment.
New Yok: Springer.

123
Ann Oper Res

Al-Shammari, M., & Masri, H. (2016). Ethical and social perspectives on global business interaction in
emerging markets. New York: IGI Global International Publishing.
Anagnostopoulos, K. P., & Mamanis, G. (2010). A portfolio optimization model with three objectives and
discrete variables. Computers & Operations Research, 37, 1285–1297.
Aouni, B., Abdealziz, F. B., & Martel, J. M. (2005). Decision-maker’s preferences modeling in the stochastic
goal programming. European Journal of Operational Research, 162, 610–618.
Aouni, B., Colapinto, C., & La Torre, D. (2014). Financial portfolio management through the goal programming
model: Current state-of-the-art. European Journal of Operational Research, 234(2), 536–545.
Aouni, B., & La Tarre, D. (2010). A generalized stochastic goal programming model. Applied Mathematics
and Computation, 215, 4347–4357.
Athanasoulis, S. G., & Shiller, R. J. (2000). The significance of the market portfolio. The Review of Financial
Studies, 13(2), 301–329.
Azmi, R., & Tamiz, M. (2010). A review of goal programming for portfolio selection. In: D. Jones et al.
(Ed.), New developments in multiple objective and goal programming. Lecture notes in economics and
mathematical systems (Vol. 638, pp. 15–33).
Ballestero, E., & PlaSantamaria, D. (2003). Portfolio selection on the Madrid exchange: A compromise
programming model. International Transactions in Operational Research, 10, 33–51.
Ballestero, E., & Romero, C. (1996). Portfolio selection: A compromise programming solution. Journal of
the Operational Research Society, 47, 1377–1386.
Ben Tal, A. (1991). Portfolio theory for the recourse certainty equivalent maximizing investor. Annals of
Operations Research, 31, 479–500.
Bereanu, B., & Peeters, G. (1970). A wait-and-see problem in stochastic linear programming: An experimental
computer code. Cahiers du Centre d’Etudes de Recherche Operationelle, 12, 133–148.
Birge, J. R., & Louveaux, F. (1997). Introduction to stochastic programming. Berlin: Springer.
Brentani, C. (2004). Portfolio management in practice. Oxford: Elsevier Butterworth-Heinemann.
Chankong, V., & Haimes, Y. Y. (1983). Multiobjective decision making: Theory and methodology. New York:
Elsevier.
Charnes, A., & Cooper, W. W. (1963). Deterministic equivalents for optimizing and satisfying under chance
constraints. Operations Research, 11, 18–39.
Charnes, A., Cooper, W., & Ferguson, R. O. (1955). Optimal estimation of executive compensation by linear
programming. Management Science, 1(2), 138–151.
Cohen, M. H., & Natoli, V. D. (2003). Risk and utility in portfolio optimization. Physica A, 324, 81–88.
Crundwell, F. K. (2008). Finance for engineers evaluation and funding of capital projects. Berlin: Springer.
Ehrgott, M. (2005). Multicriteria optimization. New York: Springer.
Ehrgott, M., Klamroth, K., & Schwehm, C. (2004). An MCDM approach to portfolio optimization. European
Journal of Operational Research, 155, 752–770.
Figueira, J., Greco, S., & Ehrgott, M. (2005). Multiple criteria decision analysis: State of the art surveys.
Berlin: Springer.
Filho, V. A. D. (2006). Portfolio management using value at risk: A comparaison between genetic algorithm and
partical swarm optimization, Master thesis informatics and Economics. Erasmus Universiteit Rotterdam.
Giannopoulos, K., Clark, E., & Tunaru, R. (2005). Portfolio selection under VaR constraints. Computational
Management Science, 2, 123–138.
Huang, X. (2008). Portfolio selection with a new definition of risk. European Journal of Operational Research,
186, 351–357.
Ida, M. (2003). Portfolio selection problem with interval coefficients. Applied Mathematics Letters, 16, 709–
713.
Jones, D. F., & Tamiz, M. (2002). Goal programming in the period 1990–2000. In M. Ehrgott & X. Gandibleux
(Eds.), Multiple criteria optimization: State of the art annotated bibliographic surveys (pp. 129–170).
Berlin: Springer.
Jorion, P. (2001). Value at risk: The new benchmark for managing risk (2nd ed.). New York: McGraw-Hill.
Kall, P., & Wallace, S. (1995). Stochastic programming. Hoboken: Wiley.
Kandasamy, H. (2008). Portfolio selection under various risk measures. Ann Arbor: ProQuest.
Lee, C. F., Finnerty, J. E., & Wort, D. H. (2010). Capital asset pricing model and beta forecasting. In Handbook
of quantitative finance and risk management (Vol. II, pp. 93–109).
Lintner, J. (1965). The valuation of risk assets and the selection of risky investments in stock portfolios and
capital budgets. The review of economics and statistics, 47(1), 13–37.
Liu, L. (2004). A new foundation for the mean-variance analysis. European Journal of Operational Research,
158, 229–242.
Mansour, N., Rebai, A., & Aouni, B. (2007). Portfolio selection through imprecise goal programming model:
Integration of the manager’s preferences. Journal of Industrial Engineering International, 3(5), 1–8.

123
 Ann Oper Res

Markowitz, H. (1952). Portfolio selection. The Journal of Finance, 7(1), 77–91.

Markowitz, H. (1959). Portfolio selection: Efficient diversification of investment. Hoboken: Wiley.
Masmoudi, M., & Ben Abdelaziz, F. (2012). A recourse goal programming approach for the portfolio selection
problem. INFOR: Information Systems and Operational Research, 50(3), 134–139.
Masmoudi, M., & Ben Abdelaziz, F. (2015). A chance constrained recourse approach for the portfolio selection
problem. Annals of Operations Research. doi:10.1007/s10479-015-1844-2.
Masri, H. (2015). A multiple stochastic goal programming approach for the agent portfolio selection problem.
Annals of Operations Research. doi:10.1007/s10479-015-1884-7.
Masri, H., Abdelaziz, F. B., & Meftahi, I. (2010). A multiple objective stochastic portfolio selection program
with partial information on probability distribution. In Computer and network technology (ICCNT), 2010
second international conference on IEEE (pp. 536–539).
Masri, H., & Abdelaziz, F. B. (2010). Belief linear programming. International Journal of Approximate
Reasoning, 51(8), 973–983.
Prakash, A. J., Chang, C., & Pactwa, T. E. (2003). Selecting a portfolio with skewness: Recent evidence from
US European, and Latin American equity markets. Journal of Banking and Finance, 27, 1375–1390.
Prékopa, A. (1995). Stochastic programming. New York: Springer.
Rachev, S., & Stoyanov, S. W. (2008). Advanced stochastic models, risk assessment, and portfolio optimization:
The ideal risk, uncertainty, and performance measures. Wiley.
Rasmussen, M. (2003). Quantitative portfolio optimization, asset allocation and risk management. New York:
Palgrave Macmillan.
Romero, C. (1991). Handbook of critical issues in goal programming. Oxford: Pergamon Press.
Roy, A. D. (1952). Safety-first and the holding of assets. Econometrics, 20, 431–449.
Sen, S., & Higle, J. L. (1999). An introductory tutorial on stochastic linear programming models. Interfaces,
29(2), 33–61.
Sharpe, W. (1964). Capital asset prices: A theory of market equilibrium under conditions of risk. Journal of
Finance, 19, 425–442.
Steuer, R., Qi, Y., & Hirschberger, M. (2005). Multiple objectives in portfolio selection. Journal of Financial
Decision Making, 1(1), 11–26.
Xia, Y., Liu, B., Wang, S., & Lai, K. K. (2000). A model for portfolio selection with order of expected returns.
Computers & Operations Research, 27, 409–422.
Xidonas, P., Askounis, D., & Psarras, J. (2009). Common stock portfolio selection: A multiple criteria decision
making methodology and an application to the Athens stock exchange. Operation Research International
Journal, 9, 55–79.
Xidonas, P., Mavrotas, G., Krintas, T., Psarras, J., & Zopounidis, C. (2012). Multicriteria portfolio management.
New York: Springer.
Xu, J., Zhou, X., & Li, S. (2011). A class of chance constrained multi-objective portfolio selection model under
fuzzy random environment. The Journal of Optimization Theory and Applications, 150(3), 530–552.
Zenios, S. A., & Ziemba, W. T. (2006). Handbook of asset and liability management. Amsterdam: Elsevier.

123