Abstract:Index tracking is a very common and popular approach in portfolio management. When there is neither (nonconvex) transaction costs nor minimal transaction unit constraints, the problem can be formulated as a convex least square problem, so that it can be solved by standard methods. However, when the transaction cost is nonconvex and not negligible, or if there is a minimal unit constraint on the amount of transaction, the problem becomes a nonconvex minimization problem with discrete variables. In this paper, w… Show more
“…The right graph shows box plots of transformed empirical distribution functions for all numbers of function evaluations. For the box plots each function value is transformed into the percentage by which it is worse than the optimal solution obtained by CPLEX Note that a problem like (2) can be formulated as a small/medium size mixed integer quadratic programming problem (MIQP), when using linearization techniques (see, e.g., Konno and Wijayanayake 2001;Guastaroba et al 2009). MIQP problems can be solved to optimality with a general purpose solver (e.g.…”
Estimation errors in both the expected returns and the covariance matrix hamper the construction of reliable portfolios within the Markowitz framework. Robust techniques that incorporate the uncertainty about the unknown parameters are suggested in the literature. We propose a modification as well as an extension of such a technique and compare both with another robust approach. In order to eliminate oversimplifications of Markowitz' portfolio theory, we generalize the optimization framework to better emulate a more realistic investment environment. Because the adjusted optimization problem is no longer solvable with standard algorithms, we employ a hybrid heuristic to tackle this problem. Our empirical analysis is conducted with a moving time window for returns of the German stock index DAX100. The results of all three robust approaches yield more stable portfolio compositions than those of the original Markowitz framework. Moreover, the out-of-sample risk of the robust approaches is lower and less volatile while their returns are not necessarily smaller.
“…The right graph shows box plots of transformed empirical distribution functions for all numbers of function evaluations. For the box plots each function value is transformed into the percentage by which it is worse than the optimal solution obtained by CPLEX Note that a problem like (2) can be formulated as a small/medium size mixed integer quadratic programming problem (MIQP), when using linearization techniques (see, e.g., Konno and Wijayanayake 2001;Guastaroba et al 2009). MIQP problems can be solved to optimality with a general purpose solver (e.g.…”
Estimation errors in both the expected returns and the covariance matrix hamper the construction of reliable portfolios within the Markowitz framework. Robust techniques that incorporate the uncertainty about the unknown parameters are suggested in the literature. We propose a modification as well as an extension of such a technique and compare both with another robust approach. In order to eliminate oversimplifications of Markowitz' portfolio theory, we generalize the optimization framework to better emulate a more realistic investment environment. Because the adjusted optimization problem is no longer solvable with standard algorithms, we employ a hybrid heuristic to tackle this problem. Our empirical analysis is conducted with a moving time window for returns of the German stock index DAX100. The results of all three robust approaches yield more stable portfolio compositions than those of the original Markowitz framework. Moreover, the out-of-sample risk of the robust approaches is lower and less volatile while their returns are not necessarily smaller.
“…A global optimal solution of the program NLP may be found by using a branch-andbound algorithm based on a decomposition of the feasible set of the program, similar to that described in Horst et al (2000); Horst and Tuy (1993); Konno and Wijayanayake (2001). This procedure exploits a binary tree, where each node is associated to a Nonlinear Program of the form .…”
Section: Branch-and-bound Algorithm For the Optimization Problemmentioning
confidence: 99%
“…As is discussed in Konno and Wijayanayake (2001), for this method to be efficient, techniques for finding lower and upper bounds have to be developed. If for a given node the current lower bound was greater than or equal to the best upper bound, then there is no need to search from this node.…”
Section: Branch-and-bound Algorithm For the Optimization Problemmentioning
confidence: 99%
“…The variable for branching the tree from the node previously chosen is chosen using an adaptation of the heuristic rule presented in Konno and Wijayanayake (2001). Let x * * = (x * * k ) k = 1,2,..., n ∈ R n be an optimal solution of the program NLP(node), associated with the current node.…”
Section: Branch-and-bound Algorithm For the Optimization Problemmentioning
This paper discusses an engineering optimization problem which arises in hydraulics and is related to the use of a new criterion for sizing water distribution piping in large buildings. The optimization model aims to find the most suitable interior pipe diameters for the various pipes in the system, using commercial sizes and minimizing the overall installation cost according to some boundary conditions. The problem is formulated as a nonconvex nonlinear program and a branch-and-bound algorithm is introduced for its solution. A procedure is proposed to obtain a feasible solution with standard values from the optimal solution of the nonconvex program. The performance of the algorithm is analysed for a real-life problem and the cost of the computed solution is assessed, showing the appropriateness of the model and the optimization techniques.
“…Fbr example, in Konno and Wijayanayake [9], they considered a rebalancing portfolio optimization problem with concave transaction cost functions and they proposed a branch and bound algorithm to solve this problem. In Konno and Yamamoto [11] [10].…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.