جستجوی مقالات مرتبط با کلیدواژه « برنامه ریزی کوادراتیک » در نشریات گروه « علوم انسانی »

Short-selling prohibition has been one of the primary assumptions of Markowitz mean-variance model. Solving Markowitz quadratic model creates investment efficient frontier by considering only two return and budget constraints. In order to develop a more realistic portfolio selection model، in this paper، a new mathematical model is developed to allow short-selling under some practical constraints. Non-linear model offered is maped by using solved standard tools and constrained efficient frontier with using from 15 shares price information.

The optimal allocation of financial resources in capital market is one of the major issues in the field of investment decisions. Making an effective decision in this regard requires the existence of appropriate investment fields, instruments and analytical techniques in the capital market. One of this efficient techniques which in addition to having unique benefits, is also recognized as the foundation of new investment strategies, is called index tracking. By considering the importance and the undeniable role of this approach in the prosperity of capital markets, its investigation and implementation has been considered in this research and so the problem of optimal TEDPIX tracker fund selection is studied via the hybrid approach of genetic algorithm and quadratic programming. We should mention that, for the data simulation and provision of the conditions needed to implement the suggested algorithm, the heuristic approach of neural network is also being used. The Results of data analysis showed the high accuracy and appropriate performance of the portfolios created by this hybrid approach in several times of iterations, so that accessing similar and better performance compared to the index could be considered as its unique characteristics.

The process of attempting to reproduce the performance of an index by investing in a subset of stocks included in it, is called index tracking. The key benefits of this approach— in risk reduction and cost control— has resulted in its tremendous evolution over the past 30 years, and no doubt it will be the foundation for the investment strategies of the future. The financial crisis and the uncertainty about the return of investments reveal the need for a low risk strategy with an appropriate return more than ever. Specifically, the current conditions in the Tehran exchange, and the ascending trend of Iran's capital market supports the theoretical basis for this investment strategy. In this context, this research pursues the problem of optimal TEDPIX tracker fund selection via the heuristic approaches introduced in the investment literature and their development using a comparative analysis for the detection of the best approach. For the data simulation and provision of the conditions needed to implement the suggested algorithms, neural network is also being used. The results showed that the high accuracy and appropriate performance of the portfolios created by multistage genetic algorithm, for accessing similar and better performance compared to the index, could be considered its unique characteristics.

The focus of this paper is on standard Markowitz mean–variance model and its traditional approach to solve portfolio selection problem (Quadratic Planning). For this goal we have applied a meta-heuristic method based on genetic algorithms (GA) in order to trace out the efficient frontier associated with the portfolio selection problem under cardinality and bounding constraints. These constraints ensure the investment in a given number of different assets and limit the amount of capital to be invested in each asset. We have presented some experimental results in two samples from Iranian stock market and overseas ones and compare the GA result with unconstrained quadratic results. Finally, we have found out which proposed GA can optimize portfolio selection problem under cardinality and bounded constrains