@ARTICLE{Rezaei, author = {Rezaei, M.H. and Ghahtarani, A. R. and Najafi, A. A. and }, title = {Application of Robust Optimization in Portfolio Selection Problem Through the Use of Conditional Drawdown at Risk}, volume = {14}, number = {2}, abstract ={Portfolio selection problem is one of the most important problems in finance. This problem tries to determine the optimal investment allocation such that the investment return be maximized and investment risk be minimized. Many risk measures have been developed in the literature until now; however, Conditional Drawdown at Risk is the newest one, which is a conditional risk value type problem. The classic model developed by this measure is a linear programming model and does not address lack of data uncertainty. In recent years many approaches have been used to consider data uncertainty; among them one of the most important and applied one is robust optimization. In a robust optimization, through using a set of uncertainties for non-deterministic constraints, a steady counterpart is defined. The present paper seeks to develop a portfolio selection model whose risk measure is to reduce Conditional Drawdown at Risk through robust optimization. The robust approach used in this research is the Bertsimas and Sim approach. In this approach, the firm counterpart provided for a linear programming model remains linear, which makes it possible to maintain the benefits of linear programming model. To test the research, we applied the model in Tehran Stock Exchange Market for 20 real shares’ data. The results showed that the model has an acceptable performance, and the results indicated the high performance of the model in developing models under uncertainty conditions. The results also showed that if the level of conservatism increases, the value of the target function will increase. }, URL = {http://jamlu.liau.ac.ir/article-1-777-en.html}, eprint = {http://jamlu.liau.ac.ir/article-1-777-en.pdf}, journal = {Journal of Operational Research and Its Applications}, doi = {}, year = {2017} }