A fuzzy approach to portfolio selection at bursa Malaysia / Wan Rosanisah Wan Mohd

Selecting the right portfolio is one of the problems to fund managers, investors, individual or institutional investors. Some authors introduced portfolio models to solve the portfolio problem such as Markowitz, Fishburn, Konno and Yamazaki, Jorion and Young. The model introduced by the authors did not consider fuzzy number in their model. The portfolio problem arises due to the uncertainty in stock market investment. Therefore, some scholars are seeking a new way to solve uncertainty of stock market investment. The fuzzy approach is the suitable approach to solve the portfolio problem. The scholars that considered the fuzzy approach are Katagiri and Ishii, Inuiguichi and Tanino, Tanaka et aI., Vercher et al. and Mohamed. In this study, we refer the extended mean-variance as a controller for our analysis purpose. The problem of the extended mean-variance model is the model assumed that the return distribution is normally distributed and the covariance is not in fuzzy numbers. Hence, the objective of this study is to determine the behaviour return data distribution and to improve the extended mean-variance model by considering the actual return distribution and fuzzy covariance. It is found the return distribution of the stock market is skewed that is not normal. Thus, the appropriate central tendency to measure the skewed return distribution in the fuzzy situation is represented by the mode. Then, the centroid was used to determine the covariance in the extended mean-variance model since the covariance has a relationship with the asset return. The data corresponds to monthly price from February 1998 until December 2011 were considered and analyzed it to the non-financial sector portfolio in Bursa Malaysia. Then, the effectiveness of our proposed model was compared with other fuzzy portfolio model that is VBS fuzzy model and extended mean-variance model. This study will give the true picture of the relationship between fuzzy return and covariance. Besides that, the application of fuzzy approach in selecting the portfolio is expected to provide useful model to the investor.