Optimal portfolio allocation using portfolio theory and heuristics driven evolutionary technique

The problem of portfolio optimization involves selecting appropriate stocks for investment by maximising the returns from the portfolio at a pre‐specified level of risk. The current approaches center around Markowitz’s mean variance optimization method that suffers from several pitfalls like instability of beta, and are either computation extensive or lead to sub‐optimal solutions. The present work suggests a heuristics and evolutionary approaches to portfolio optimization. The approach is computationally less intensive. It further extends the approach to include cardinality constraints that limit a portfolio to have a specified number of assets, and to impose limits on the proportion of the portfolio held in a given asset. The heuristics technique is employed for asset selection while the evolutionary technique is used for allocation of funds among the already selected assets. The approach is capable of handling a large number of instruments and scenarios, and is relatively stable to minor variations of the inputs, as is practiced in real life situations. The performance from this approach compares well with the Markowitz’s model, and performs better than the stock market indices of US and India.