| The wealth of an individual can consist only of cash or a combination of cash and securities. If it consists only of cash, the problem involved in optimally allocating it into a collection of securities is known as portfolio selection. If it consists of cash and securities the problem is one of revising. The difference arises due to taxes and transaction costs. Studies in modern portfolio theory have emphasized portfolio selection rather than portfolio revision. This study emphasizes portfolio revision as an important problem in practical applications.;This study suggests two procedures to solve the problem of portfolio revision. In both methods pairwise comparisons of securities is made to choose securities for substitution at the margin. This procedure of substitution is consistent with the Markowitz principle of diversification, as a security is not considered in isolation, but for its contribution to the portfolio in terms of the mean-variance characteristics of the portfolio.;A notable feature of the suggested models is that when substitutions are made (at the margin) in each iteration, the actual trading of securities is simulated. That is, using the current prices of securities, transaction costs, and taxes involved, the sale and purchase of securities are simulated. This aspect of the model is of particular importance to investors in practical applications. Pairwise substitutions of securities can result in suboptimal solutions. In the context of portfolio revision this shortcoming is more than compensated by the benefits associated with it. In this framework, not only taxes and transaction costs are incorporated, but portfolio turnover can also be included in the analysis. If formulated as a general non-linear programming problem it is of no practical significance as it cannot be used for solving problems of the size usually encountered. The suggested procedures, in contrast, with some modifications, will provide acceptable solutions.;A comparison of the existing literature in the area of portfolio selection and portfolio revision reveals that portfolio selection models are formulated such that optimal results are obtained. In contrast, works on portfolio revision emphasize obtaining workable solutions which are usually suboptimal. |
