Robust covariance estimators for mean-variance portfolio optimization with transaction lots

Rosadi, Dedi; Setiawan, Ezra Putranda; Templ, Matthias; Filzmoser, Peter Article Robust covariance estimators for mean-variance portfolio optimization with transaction lots Operations Research Perspectives Provided in Cooperation with: Elsevier Suggested Citation: Rosadi, Dedi; Setiawan, Ezra Putranda; Templ, Matthias; Filzmoser, Peter (2020) : Robust covariance estimators for mean-variance portfolio optimization with transaction lots, Operations Research Perspectives, ISSN 2214-7160, Elsevier, Amsterdam, Vol. 7, pp. 1-11, https://doi.org/10.1016/j.orp.2020.100154 This Version is available at: https://hdl.handle.net/10419/246426 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Contents lists available at ScienceDirect Operations Research Perspectives journal homepage: www.elsevier.com/locate/orp Robust covariance estimators for mean-variance portfolio optimization with transaction lots Dedi Rosadi a , Ezra Putranda Setiawan b , Matthias Templ c , Peter Filzmoser ⁎,d a Department of Mathematics, Universitas Gadjah Mada, Indonesia b Department of Mathematics Education, Universitas Negeri Yogyakarta, Indonesia c Institute of Data Analysis and Process Design, Zurich University of Applied Sciences, Winterthur, Switzerland d Institute of Statistics and Mathematical Methods in Economics, TU Wien, Austria ARTICLE INFO Keywords: Finance Markowitz portfolio Transaction lots Robust estimation Genetic algorithm ABSTRACT This study presents an improvement to the mean-variance portfolio optimization model, by considering both the integer transaction lots and a robust estimator of the covariance matrices. Four robust estimators were tested, namely the Minimum Covariance Determinant, the S, the MM, and the Orthogonalized Gnanadesikan–Kettenring estimator. These integer optimization problems were solved using genetic algorithms. We introduce the lot turnover measure, a modified portfolio turnover, and the Robust Sharpe Ratio as the measure of portfolio performance. Based on the simulation studies and the empirical results, this study shows that the robust estimators outperform the classical MLE when data contain outliers and when the lots have moderate sizes, e.g. 500 shares or less per lot. 1. Introduction For several decades, trading activities in several stock exchanges have increased, since more investors buy assets and sell them again or keep them until the dividend is shared. However, the investors are always facing the risk that the asset price will move downward. To reduce this type of risk, the investor can diversify by buying different assets. Therefore, a loss in an asset can be covered by profits of other assets. To determine the capital allocation in a portfolio of stocks, Markowitz [24] introduced the mean-variance model. In this model, the variance was used as a risk measure, and the goal was to minimize the variance of the portfolio. As has been noted, e.g. in Britten-Jones [5], the Markowitz model became important, popular, and widely used in practice. In the Markowitz model, the classical mean and covariance matrix estimators are used. These estimators rely on the assumption of multivariate normal distributed returns, which is rarely fulfilled in real applications. Lauprete et al. [19] showed that many empirical portfolio returns have the sample skewness and the sample kurtosis which exhibit fat tails, follow a non-symmetric distribution and have multivariate tail dependence. Robust estimators can be applied to handle data that contain outliers and deviate from the assumption of multivariate normality. Huber and Ronchetti [14] characterized the robustness of estimators in terms of their sensitivity to small deviations from the assumptions of the underlying distribution of the data. The properties of robust estimators have been studied extensively and described in the literature, for example, see Hampel et al. [12], Huber and Ronchetti [14], and Marrona et al. [25]. The usage of robust estimators in the portfolio optimization problem dates back to Lauprete et al. [19], who examined how the Gaussian assumption in the Markowitz mean-variance model is influenced by marginal heavy tail. Lauprete et al. [19] also prove that robust alternatives to the classical variance estimator have lower risk of loss than the non-robust ones. Perret-Gentil and Victoria-Feser [27] use robust Sestimators to estimate the mean and variance in Markowitz’s portfolio optimization model. Welsch and Zhou [37] consider several robust covariance estimators in the mean-variance portfolio optimization, such as FAST-MCD, Iterated Bivariate Winsorization, and Fast 2DWinsorization. DeMiguel and Nogales [9] prove that certain robust estimators produce more stable and less sensitive portfolios than the traditional portfolio. Kaszuba [18] compares several robust estimators that are used in the mean-variance portfolio optimization model and the traditional mean-variance portfolio. He finds that the risk of MPortfolios and LAD-Portfolios is significantly different from the risk of a classical portfolio. Recently, Supandi et al. [34] show that portfolios that use constrained M-estimators are better than the classical portfolio. It must be noted that the term ’Robust Portfolio Optimization’does not https://doi.org/10.1016/j.orp.2020.100154 Received 8 December 2019; Received in revised form 6 June 2020; Accepted 6 June 2020 ⁎ Corresponding author. E-mail addresses: [email protected] (D. Rosadi), [email protected] (E.P. Setiawan), [email protected] (M. Templ), peter.fi[email protected] (P. Filzmoser). Operations Research Perspectives 7 (2020) 100154 Available online 15 June 2020 2214-7160/ © 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/). T only refer to the usage of robust estimators in portfolio optimization. The usage of robust optimization techniques for solving the portfolio optimization problem (see e.g. Goldfarb and Iyengar [11] is also known as robust portfolio optimization. On the other hand, all stock markets around the world have determined a minimum number of shares or assets that can be traded regularly, known as transaction lot. For example, in the Indonesia Stock Exchange, one lot of stocks consists of 100 shares, so the investor can trade 100, 200, etc. shares regularly. Since the portfolio weights obtained by the Markowitz model were presented as fractions or percentages to the total capital, they could not be applied directly by investment managers. The portfolio optimization problem by the Markowitz model with the transaction lot constraint has been discussed in the literature, for example, by [20,23], and Buchheim et al. [6], Mencarelli and D. Ambrosio [26], Soleimani et al. [32], including the review paper in Mansini et al. [22]. Other studies with the transaction lot constraint, but using different risk measures are available, for instance, Setiawan and Rosadi [30] which uses the Value at Risk; Barati et al. [4] which uses the semi variance, etc. However, all of these studies only used the classical variance estimator with additional integer lots constraint and did not consider the robustness of their estimator. In this paper, we present an improvement to the mean-variance portfolio optimization model with the integer transaction lots constraint, by considering robust estimators of the covariance matrices to deal with the presence of outliers in the data. Four robust estimators are tested, namely the Minimum Covariance Determinant (MCD) estimator, the Tukey’s Biweight S-estimator, the Orthogonalized GnanadesikanKettenring (OGK) estimator, and the MM-estimator. To obtain the integer solution to the portfolio optimization problem, we use a genetic algorithm (GA), see, e.g., Arnone et al. [1], Soleimani et al. [32] and Chang et al. [8]. Several nature-inspired metaheuristics alternatives or the (mixed) integer programming methods can also be considered to solve this problem (which is a subject of future research), see, e.g., Bacanin and Tuba [2],3], Lubis et al. [21], Sitopul et al. [31], Strumberger et al. [33], Tuba and Bacanin [36] and the references therein. Therefore, this study can be viewed as a mixture of robust statistics based and transaction lots based portfolio optimization. We also introduce the lot turnover measure, a modified portfolio turnover, and the Robust Sharpe Ratio as the measure of portfolio performance. Based on the simulation studies and the empirical results, this study shows that the robust estimators outperform the classical MLE when data contain outliers and when the lots have moderate sizes. This paper is organized as follows. In Section 2, we discuss the theoretical formulation of the mean-variance portfolio optimization and present various robust estimators. In Section 3, we present the mean-variance portfolio optimization problem with transaction lots constraint and the robust GA approach solving the problem. In the next section, Section 4, we present simulation studies whereas in Section 5 we provide the empirical results using stocks from Indonesia Stocks Exchange (IDX). The last section is devoted to the conclusion and some remarks on directions of future research. 2. Robust statistics in the portfolio optimization model In recent decades, robust statistical methods have been used in the portfolio optimization model, i.e. •Robust risk measures have been used to substitute the variance in portfolio optimization. Instead of the variance, the portfolio’s risk is measured by the Least Absolute Deviation (LAD) or by other types of M-estimators, as presented e.g. in Kaszuba [18] and Fabozzi et al. [10]. According to Kaszuba [17], this method is known as a one-step approach. •Robust methods have been used to estimate the covariance matrix in the Markowitz mean-variance portfolio optimization model. This model has been discussed in several papers, for example, Kaszuba [18], Welsch and Zhou [37], and Supandi et al. [34]. There are many robust estimators of the covariance matrix, for example, the Minimum Covariance Determinant (MCD) estimator, the Constrained M-estimator, etc. According to Kaszuba [17], this method is also known as a two-steps approach. In the following subsection, we discuss the Markowitz portfolio optimization model and some robust statistics that can be used to estimate the covariance matrix in our modified Markowitz model. 2.1. Markowitz’s portfolio optimization model Suppose that there are nrisky assets traded in the stock market. Define ==…wi nw [], 1,2, , i as a column vector representing the portfolio weights, and ==… μ μi n[], 1,2, , , ia column vector representing the expected returns of each asset. The expected returns for this portfolio are Fabozzi et al. [10] ∑ == = μwμμw pT i n ii 1 To calculate the portfolio’s variance, one should calculate the variance-covariance matrix =⎛ ⎝ ⎜ ⎜ … … ⋮⋮⋱⋮ … ⎞ ⎠ ⎟ ⎟ σσ σ σσ σ σσ σ Σ. n n nn nn 11 12 1 21 22 2 12 (1) The portfolio’s variance is defined as ∑∑ == == σ wwσwΣw pT i n j n ijij 2 11 (2) By assuming that short-selling is prohibited, the Markowitz portfolio optimization problem can be stated as =σwΣwmin wpT 2 subject to 1. =μμ w T0 2. = w 1 1 Tp 3. w i ≥0, for =… i n1, 2, , . Here, μ 0 ∈Rdenotes a constant. To solve this optimization problem in practice, one needs to know the covariance matrix Σand the mean vector μ, which are usually estimated from the historical data of the asset returns. Let y t· represent an n-dimensional column vector containing the data of the sample returns of all nassets at time t, namely =… t T1, 2, , . Suppose the historical data follow a multivariate normal distribution, then the maximum likelihood estimation for each parameter is ∑ = = μ Ty ^1 t T t 1 ·(3) and ∑ =−− = μμ T Σ yy ^1(^)( ^). t T tt T 1 ·· (4) As noted by Perret-Gentil and Victoria-Feser [27], both equations above are the most efficient estimators, under the assumption of multivariate normality of the data. 2.2. Robust covariance estimators Statistical procedures often depend on one or several assumptions about the (probability) distribution of the data. Fabozzi et al. [10] state D. Rosadi, et al. Operations Research Perspectives 7 (2020) 100154 2 that robust statistics addresses the problem of making estimates that are insensitive to small deviations from the basic assumption of the statistical models employed. In this section, we review some well-known robust estimators of the mean and the covariance matrix. We describe several robust estimators that can withstand a high fraction (up to 50%) of outliers, such as the Minimum Covariance Determinant (MCD) estimator, S-estimators and MM-estimators. We also discuss a method suitable for high-dimensional data, namely the Orthogonalized Gnanadesikan-Kettenring (OGK) estimator. 2.2.1. Orthogonalized Gnanadesikan-Kettenring estimator The OGK estimator is a robust estimator of covariance that is symmetric but not necessarily positive definite, nor affine equivariant. According to Marrona et al. [25], the computation of the OGK estimator is as follows. As in the previous sub-section, let =y Y [] ti be a T×ndata matrix, where the rows y t· denote the returns of each stock at time t, =… t T1, 2, , . Hence, the columns y ·i denote the changes of the returns for each stock, =… i n1, 2, , . 1. Compute a normalized data matrix Xwith columns =σ xy y ^() , ii i ·· · where σ ^ is a univariate robust estimator of variance. Consequently, the rows will be: ==… −tTxAy,1,2,, , tt ·1· where =…σσAyydiag(^(),, ^()) . n·1 · 2. Compute a robust correlation matrix U, with elements =U 1 ii and =+−−Uσ σxx xx 1 4[^() ^()], ij ij ij·· 2·· 2 for i≠j. 3. Compute the eigenvalues of Uand the corresponding eigenvectors. Let Ebe the matrix where the columns are the eigenvectors of U. Then, compute the matrix Zwith the rows == =… −tTzExEAy,1,2,,. tTtTt ·· 1· 4. For each =… i n1, 2, , , compute μ z ^( ) i· and σ z ^() , i·where μ ^ is a robust univariate location estimator. Then, set =…σσΓzzdiag(^(),, ^()) n·1 2·2 and =…νμμzz(^(),, ^()) n T ·1 · 5. We obtain the Orthogonalized Gnanadesikan-Kettenring estimator for the (multivariate) mean as = μν YAE ^() ( ) and for the covariance matrix as, = Σ YAEΓAE ^()()() . T 2.2.2. Minimum covariance determinant estimator The Minimum Covariance Determinant (MCD) estimator is a robust estimator of location and covariance, introduced by [28]. This estimator uses both mean and covariance matrix of hdata points, for ≤<hT , T 2 with the smallest determinant of the population covariance matrix. The mean is estimated by ∑ = = μ hy ^1 MCD t h t 1 · and the estimated covariance is ∑ =−− =μμ h Σ yy ^1(^)( ^) MCD t h tMCDtMCD T 1 ·· It is known that the breakdown value of this estimator equals −Th T. The exact MCD estimator is very hard to compute, as it requires the evaluation of all ⎜⎟ ⎛ ⎝⎞ ⎠ n h subsets of size h. To simplify and to speed up the computation, one can implement the FAST-MCD algorithm, as proposed by [29]. The FAST-MCD algorithm is as follows: 1. Randomly draw hobservations from Tavailable data, where ≤<hT T 2. With this subset, compute the empirical mean and covariance matrix, namely μ ^0and Σ ^ 0 . Note that if =Σdet(^)0 , 0 add one or more observation until ≠Σdet(^) 0 0 . 2. For each observation =… t T1, 2, , , compute the distance as =− − − μμ d tyΣy() ( ) ^( ) tTt 0 2·0 0 1 ·0 and set = Σ Σ ^:^ old 0. 3. Sort the distances from the smallest to the largest one, i.e. ≤≤…≤ d πdπ dπT((1)) ((2)) (()) , 00 0 where π(t) indicates the permutation to obtain the ordering. 4. Get the new subset of size has the observations corresponding to the smallest hdistances, namely the index set =…Hππ πh{ (1), (2). , ( )} 1. 5. In the new subset, compute the new mean and covariance matrix: μ ^0 and Σ ^ 0 . 6. Since ≤ΣΣdet(^)det( ^), old0repeat steps 2–5 until =ΣΣdet(^)det( ^) old0. 7. Repeat the step 1–6 for several initializations. 8. The last μ ^0and Σ ^ 0 are the MCD estimates. In step 7, the algorithm will stop and obtain an approximate MCD solution by taking several initial choices of h, applying steps 1–7 to each random choice of hobservation, and keeping the solution with the lowest determinant. Since there is only a finite number of hsubsets, the sequence of determinants calculated by this algorithm must converge within a finite number of steps. Note that there is no guarantee that the final iteration will give the global minimum of the covariance determinant. Another algorithm to calculate the MCD estimator for the mean and the covariance matrix has been suggested by Hubert et al. [16], known as the deterministic MCD algorithm. 2.2.3. S-estimator The S-estimator could be recognized as a generalization of the Minimum Volume Ellipsoid (MVE) estimator, to increase the efficiency of the MVE multivariate location and scatter estimator. According to Marrona et al. [25], the S-estimators of location μ ^ S and covariance Σ ^ S are defined such that the determinant of the matrix Σ ^ S is minimized under the constraint ∑ ⎜⎟ ⎛ ⎝−− ⎞ ⎠= = − μμ TρbyΣy 1(^)^(^) , t T tS TStS 1 · 1 · where bis a constant, ∈ μ ^R , Snand Σ ^ S is a positive definite symmetric matrix. Here, the loss function ρ(.) must be carefully selected from the continuously differentiable function class, such that the S-estimator will have a high breakdown and is asymptotically normal. A popular choice of continuously differentiable loss functions (which is also used in this paper) is Tukey’s biweight function [15], given by D. Rosadi, et al. Operations Research Perspectives 7 (2020) 100154 3 =⎧ ⎨ ⎪ ⎩ ⎪ −⎡ ⎣−⎤ ⎦≤ > () () ρ u um um () 11 ,for ,for. mu m m 6 23 6 2 The breakdown value of Tukey’s biweight S-estimator depends on the constant m, i.e. = ϵ * b m 6. Other ρfunctions could be used as well, for example, Hampel’s function or Huber’s function Hubert and Rousseeuw [15]. 2.2.4. MM-estimator The MM-estimator Tatsuoka and Tyler [35] is a type of high breakdown value estimator, an extension of the S-estimator, that has high efficiency under multivariate normality. The computation of the MM-estimator for the returns data =…tTy,1,2,, , t·consists the following four steps: 1. First, choose a loss function ρto compute the S-estimators of location and covariance, namely μΣ ( ˜,˜ ) . 2. Compute = σ Σ ^˜n 1 2. 3. Find the MM-estimator of the location and the shape parameter, μΓ ( ^,^), MM that minimize ∑⎛ ⎝ ⎜−− ⎞ ⎠ ⎟ = − μμ Tρσ yΓy 1[( ) ( )] ^ t T tTt 1 1 ·1· 1 2 for all μ∈R n and =∈PDS nΓΓ1, ( ) , the class of positive definite symmetric matrices. Here ρ 1 does not need to be the same as ρ. 4. Compute the MM-estimator of covariance matrix, = Σ σ Γ ^^^ MM 2. The breakdown point of the MM-estimator is inherited by that of the S-estimator used for its computation Hubert and Rousseeuw [15]. 3. Minimum transaction lots in the Markowitz model Most of the financial asset transactions take place at the stocks market, where companies offer their stocks, bonds, or derivative shares. A minimum number of shares that can be regularly traded in the stock markets is called a transaction lot. As an example, for regular transactions at the Indonesia Stocks Exchange (IDX), one transaction lot for stocks and options contract is 100 units and 1 unit, respectively. The number of asset lots that can be traded by an investor is represented as an integer. 3.1. Optimal portfolio with transaction lots and robust covariance estimator Suppose that bis the amount of capital that will be invested into n available shares at time t. Suppose that the price per share of the ith shares is c i , and the number of shares per lot is k.IfΣdenotes the variance-covariance matrix (estimated using previous data), we can determine the weight of the ith asset, w i as well as the number of lots of the ith asset, x i , by solving the following optimization problem (assuming that short sale is prohibited): wΣwmin T subject to 1. ∑ ≤ =kx c b i nii 1 2. ==… ∑ = w jn,1,2,, . j cx cx jj i nii 1 3. x j is an integer. It must be noted that the sample mean would contribute a high estimation error to the out-of-sample performance DeMiguel and Nogales [9], Kaszuba [18]. Therefore, following Kaszuba [18], our model only considers the minimum-variance portfolios, which are obtained by removing the constraint of the sample mean. Hence, the optimal portfolio weights only depend on the covariance matrix. The first constraint in the optimization above ensures that the total money spent to buy the stocks does not exceed the available capital (b), while the last constraint confirms that the lot numbers are integers. In the practical computation, the covariance matrix is replaced by its estimates. Besides the classical maximum likelihood estimation (MLE), in this study, we apply four different robust estimators to estimate the Σ, namely the MCD estimator, the OGK estimator, the S-estimator, and the MM-estimator. On the other hand, the above mentioned portfolio minimum variance selection with transaction lots is a multi-objective decision problem with integer feasible regions. This problem can be solved using the genetic algorithm method. Other alternatives are available, such as the (mixed) integer programming method or other heuristic methods, see e.g, Bacanin and Tuba [2], Lubis et al. [21] and the references therein. 3.2. Solving the optimization using the genetic algorithm One of the most popular heuristic technique that is widely used in portfolio optimization is the genetic algorithm (GA). Proposed by Holland [13], the GA had been developed from the concept of natural selection and genetic theory. In the GA, each solution candidate must be represented as chromosomes. To initialize the algorithm, an initial population that consists of several chromosomes is generated. Afterwards, several pairs of chromosomes as ’parents’are selected from the population, which produce new chromosomes (as the next generation) by crossover and/or mutation operator with predefined probability. A fitness function, which is defined based on the objective function of the problem, is used to evaluate the performance of each solution represented by each chromosome. These processes have to be repeated for several generations until the final (termination) condition has been reached. As explained in Chang et al. [7], the basic steps of the genetic algorithm are as follows. 1. Generate an initial population consisting of several chromosomes. 2. Evaluate the fitness of each chromosome in the population. 3. Select ’parents’from the population. 4. Recombine parents to produce the next generation, using crossover and mutation. 5. Evaluate the fitness of the next generation. 6. Replace some or all populations by the next generation. 7. Repeat steps 3–6 until a satisfactory solution has been found. The application of the genetic algorithm to solve the portfolio optimization problem was introduced by Arnone et al. [1]. Several studies, for example Chang et al. [7], Soleimani et al. [32], and Chang et al. [8], showed that the genetic algorithm could efficiently find a near optimal or even the optimal solution of a portfolio optimization problem. Following Lin and Liu [20], in this paper chromosomes with nelements or genes were used, where nis equal to the number of stocks in the portfolio. Each gene is a real number between 0 and 1. Let =… u j n ,1,2,, j be the jth genes in a chromosome. Since the genes should represent the portfolio weights, we must divide each number by the total sum of genes in the chromosome, =∑=… = vu ujn,1,2,, . j j j nj 1 Suppose that the price of the jth asset is c j and the total investment capital is b, while the number of lots for each asset could be calculated as =⎢ ⎣ ⎢⎥ ⎦ ⎥=…xb v cjn,1,2,, j j j The calculation of the lots of each asset would cause the change of D. Rosadi, et al. Operations Research Perspectives 7 (2020) 100154 4 the portfolio weight. Hence we define the modified weight after lot calculation, that is, =∑=… = w xc xc jn,1,2,, . j jj j njj 1 The modified weight, as part of the weight vector w, is used in the fitness function: ∑ =−− = FbxcwΣw ϵ T j n jj 1 In this function, the second term is a penalty to ensure that the amount of money is not far from the total available capital b. By adjusting the ϵvalue, the algorithm can find the portfolio with total price as near as possible to b. In other words, the investor may spend most of all available capital. 3.3. Measuring the portfolio performance: rolling horizon and the lot turnover measure Following DeMiguel and Nogales [9] and Supandi et al. [34],to evaluate the performance of a robust estimation method in portfolio optimization, we use the rolling horizon method. This method consists of the following steps: 1. Obtain the data of nstock for the period T. Determine the length of the estimation window η, where η<T. 2. Compute the portfolio assets weights wand/or the number of asset lots xbased on the data in the estimation window. To implement this step, estimate the new covariance matrix for each estimation window using both the classical and the robust estimation method. 3. Hold the portfolio (obtained in the second step) and determine the out-of-sample returns data + rt 1 . 4. Change the estimation window by including one more recent data point, and discard one the earliest data point. 5. Repeat steps 2, 3, and 4 until the end of the data set is reached. The out-of-sample returns data + r ^ t 1 obtained in the third step could be used to evaluate the portfolio performance, that is, by calculating the out-of-sample mean returns ∑ =−=+ μ Tη r 1^ , s tη T t 1 and the standard deviation ∑ =−− − =+ σ Tη rμ 1 1(^) stη T ts 1 2 We also consider the Sharpe Ratio (SR) or excess returns to risk ratio, that is, =− SR μr σ ssf s where r f represents the risk-free investment returns. Since the μ s and σ s used in the above formula are not robust, we introduce a new robust ratio measure, called Robustified Sharpe Ratio (RSR), as =− R SR μr σ s αs f αs , , while μ α,s and σ α,s denote the trimmed mean and the trimmed standard deviation, respectively. For these ratios, a higher Sharpe Ratio or Robust Sharpe Ratio indicates better portfolios, since it means that more returns are obtained with the same risk, or, the portfolio has a smaller risk for the same returns. In addition, to measure the portfolio’s stability, DeMiguel and Nogales [9] defines the portfolio turnover as ∑∑ =−− − == −+ P TTη ww 1 1, i n tη T ti t i 1 1 ,1, where w t, i and + w t i 1, denote the ith asset weight at time tand time + t 1 , respectively. In our research, since the result of portfolio optimization is presented as the number of lots, we modify the above mentioned formula to represent the average change in the number of lots, that is, ∑∑ =−− − == −+ LT Tη xx 1 1i n tη T ti t i 1 1 ,1, where x t,i and + xt i 1, denote the number of lots of the i-th asset at time t and + t 1 , respectively. Denote p ti as the price of the assets iat time t.Bydefining =∑= w xp xp ˜ti ti ti i nti ti , 1 and using the result w ˜ti,to replace the w t, i in the formula of the portfolio turnover, we can obtain the modified portfolio turnover (MPT) that could be compared with the portfolio turnover (PT). The portfolio turnover, lot turnover, as well as the modified lot turnover, represent the rebalancing of assets in the portfolio that must be done to maintain the portfolio at optimal conditions. Rebalancing assets need some money to buy the assets and pay the transaction cost, hence the better portfolio is the one that exhibits a smaller turnover. 4. Simulation study In this section, we study the property of the modified mean-variance portfolio optimization problem with additional transaction lots constraint and robust estimators for the covariance matrix via simulation studies. Here we consider a mixture of a multivariate normal distribution F δ (r) with contamination. In this model, δ∈(0, 1) represents the proportion of contamination to the multivariate normal model. Hence, the actual asset returns follow the following distribution, =− +μFr δN δHΣ() (1 ) ( , ) , δp where N p (μ,Σ) denotes the multivariate normal distribution with mean vector μand covariance matrix Σ. Following the simulation study of Supandi et al. [34], a set of returns data for three independent assets is generated. The generated data are multivariate normally distributed, with mean vector = μ (0.12, 0.15, 0.13 ) and covariance matrix Σ, where we consider the case of the independent returns with =⎛ ⎝ ⎜ ⎞ ⎠ ⎟ Σ 0.012 0 0 0 0.015 0 0 0 0.016 (5) and the dependent returns with Table 1 The parameterization of the genetic algorithm. Method (Parameters) Value Chromosome length 6 (equal to the number of assets in the portfolio) Representation of genes real number between 0 and 1 Number of population 150 Crossover method Uniform crossover Crossover probability 0.7 Mutation method Random mutation Mutation probability 0.1 Selection method Tournament (roulette-wheel) selection Elitism 3 Number of generation 2000 D. Rosadi, et al. Operations Research Perspectives 7 (2020) 100154 5 Fig. 1. The fitness function values of all populations for each generation of the genetic algorithm (in the case of independent returns with lot constraint). In the right panel, we zoom into the plot by only showing the best values of the fitness function. It clearly shows that after 2000 generations, the results are already stable. Fig. 2. Risk of portfolios calculated with transaction lots (TL) (with size 100 and 500 shares per lot) and without transaction lots (NTL) based on classical MLE and several robust estimators (independent returns). D. Rosadi, et al. Operations Research Perspectives 7 (2020) 100154 6 =⎛ ⎝ ⎜⎞ ⎠ ⎟ Σ 0.012 0.001 0.001 0.001 0.015 0.001 0.001 0.001 0.016 . (6) For the presence of any outliers, we assume that His multivariate normal distributed with the same covariance matrix Σbut with different mean vector, that is, =−−− μ (0.12, 0.15, 0.13) . H For each of these covariance matrices Σand mean vector μ H , three sets of data were generated: (1) data without outliers, (2) data with 5% outliers, and (3) data with 10 percent outliers. Data from each type Fig. 3. Risk of portfolios calculated with transaction lots (TL) (with size 100 and 500 shares per lot) and without transaction lots (NTL) based on classical MLE and several robust estimators (dependent returns). Table 2 The mean of Sharpe Ratio and Robust Sharpe Ratio (in italic) from simulated data (independent case). Estimator No Transaction Lots Lot =100 Lot =500 Type 0 pct. 5 pct. 10 pct. 0 pct. 5 pct. 10 pct. 0 pct. 5 pct. 10 pct. MLE 4.2232 4.2208 4.2285 4.2420 4.2372 4.2449 4.2434 4.2331 4.2377 4.5132 4.5108 4.5191 4.5309 4.5259 4.5345 4.5319 4.5227 4.5271 MCD 4.2531 4.2481 4.2532 4.2441 4.2393 4.2452 4.2425 4.2359 4.2423 4.5418 4.5369 4.5425 4.5337 4.5285 4.5346 4.5312 4.5255 4.5315 S 4.2530 4.2480 4.2532 4.2421 4.2394 4.2467 4.2416 4.2348 4.2414 4.5418 4.5368 4.5424 4.5311 4.5284 4.5359 4.5303 4.5245 4.5312 OGK 4.2530 4.2481 4.2532 4.2428 4.2387 4.2464 4.2411 4.2351 4.2391 4.5417 4.5369 4.5425 4.5318 4.5278 4.5357 4.5297 4.5245 4.5285 MM 4.2531 4.2483 4.2533 4.2466 4.2400 4.2438 4.2375 4.2357 4.2399 4.5418 4.5371 4.5426 4.5354 4.5292 4.5333 4.5262 4.5249 4.5293 D. Rosadi, et al. Operations Research Perspectives 7 (2020) 100154 7 were generated 200 times, meaning that we simulated returns data for three stocks in 200 observations. By assuming that the price of each stock at the beginning of the period is = P 1000 , 0the price at the end of the period can be calculated using the sum of the generated log returns. This simulation was repeated 200 times, yielding 200 different data sets. For each generated data set, the portfolio optimization problem with transaction lots (we only consider lot size 100 and 500 shares here) and without transaction lots were calculated as described in Section 3. The covariance matrix was estimated using the mentioned methods: the MLE, the MCD estimator, the S-estimator, the OGK estimator, and the MM-estimator. Based on the five covariance estimates, the optimum weights in the portfolios are obtained using the genetic algorithm approach (see Sections 3.1. and 3.2), assuming that short selling is prohibited. The parametrization for the genetic algorithm is presented in Table 1. We generate three different optimal portfolios and compare the fitness function’s value for each combination. Only portfolios with the best (largest) fitness were considered and further analyzed. Fig. 1 shows the fitness function values (the mean, the median and the best) of all populations for each generation in the genetic algorithm (in the case of independent returns with lot constraint). From Fig. 1 (right panel), we clearly can see that the results already stabilize after 2000 generations. Figs. 2 and 3show that the portfolio optimization method without transaction lot constraint based on the robust estimators has lower risk compared to the MLE risk. When the transaction lots were applied, the classical MLE and several robust estimators yield portfolios with similar risk. Tables 2 and 3summarize the mean of both the Sharpe Ratio (SR) and the Robust Sharpe Ratio (RSR) of the independent and dependent returns, respectively, as defined in Section 3 above. For the optimal portfolio without transaction lots, it can clearly be seen that the robust estimator yields better portfolios than the classical MLE in the presence of outliers, both for the independent and dependent returns. In this case, the MM-method seems to always perform better than the other robust methods. The simulation results obtained for the optimal portfolio with transaction lots can be summarized as follows. In case of Table 3 The mean of Sharpe Ratio and Robust Sharpe Ratio (in italic) from simulated data (dependent case). Estimator No Transaction Lots Lot =100 Lot =500 Type 0 pct. 5 pct. 10 pct. 0 pct. 5 pct. 10 pct. 0 pct. 5 pct. 10 pct. MLE 3.4719 3.4669 3.4675 3.4452 3.4376 3.4286 3.4272 3.4229 3.4071 3.7145 3.7106 3.7096 3.6839 3.6779 3.6673 3.6650 3.6590 3.6458 MCD 3.4794 3.4762 3.4745 3.4408 3.4331 3.4391 3.4239 3.4194 3.4165 3.7226 3.7208 3.7172 3.6793 3.6745 3.6798 3.6611 3.6556 3.6548 S 3.4799 3.4759 3.4744 3.4466 3.4342 3.4434 3.4267 3.4294 3.4101 3.7232 3.7204 3.7171 3.6849 3.6750 3.6834 3.6643 3.6663 3.6495 OGK 3.4793 3.4755 3.4739 3.4429 3.4392 3.4347 3.4361 3.4322 3.4185 3.7225 3.7200 3.7165 3.6817 3.6800 3.6737 3.6747 3.6691 3.6571 MM 3.4797 3.4762 3.4750 3.4413 3.4439 3.4391 3.4234 3.4222 3.4156 3.7230 3.7209 3.7178 3.6805 3.6849 3.6782 3.6607 3.6582 3.6552 Table 4 Price range and descriptive statistics of each asset returns. Stock Stock price Stock returns Code Min Max Mean Median Std.Dev MAD Skewness Kurtosis S1 3500 8375 0.046 0.00 0.022 0.019 0.066 3.311 S2 5050 9250 −0.013 0.00 0.021 0.017 0.069 5.226 S3 3450 9950 0.001 0.00 0.020 0.017 0.216 5.331 S4 3175 8250 0.008 0.00 0.019 0.015 0.102 5.769 S5 3100 7500 0.073 0.00 0.019 0.013 0.196 4.185 S6 4680 9200 0.019 0.00 0.021 0.014 0.102 5.049 Fig. 4. Normal QQ-plot of stock S 1 until S 6 returns data. D. Rosadi, et al. Operations Research Perspectives 7 (2020) 100154 8