Ebook A Comparative Application of Alternative DEA Models in Selecting Efficient Large Cap Market Securities in India
Stock selection to a portfolio has always been a major task. The Value Line Survey on stock selection gives a three-step approach for investors, which includes the determination of risk tolerance, picking of stocks with acceptable safety ranks with higher dividend yields and balancing of portfolio in accordance with their performance. However, the investors’ selection process may not end there since the availability of number of securities with multiple input and multiple output data makes it complex to determine efficient stocks. The stock selection and portfolio construction processes used by the non-quantitative investors can be enhanced with quantitative methodologies along with the risk models. Many investors therefore entrust the job with mutual funds. Especially, in India, this can been noticed from the increase in the assets under the mutual fund industry from Rs.121,805 crores in January 2003 to Rs.326,388 crores in March 2007. (Note that 1 crore = 10 million and Rs.: Rupees (Indian currency).
Traditionally, the investors use the Economic, Industry and Company (EIC) analysis approach. EIC approach requires a combination of both qualitative and quantitative analysis. King (1966) observed that one half of the variables in stock price could be attributed to market influence that affected all the stock market indexes, and on the average, about 13 per cent of the variation in stock price due to industry influence. Apart from the economic and industrial analysis, the company analysis also plays a major role in deciding upon the inclusion of a particular stock.
Company analysis focuses more on the chemistry of earnings, which includes earnings relationship with asset productivity, debt financing, dividends per share and price earnings ratio. Hence the variables like earnings per share (EPS) and price to earnings (P/E ratio) are considered to have influence on the investors’ decisions. Therefore forecasting models have always been useful in this venture.
The research on stock selection process in recent years uses various quantitative methods such as data envelopment analysis (DEA), artificial intelligence (AI), artificial neural networks (ANN) and Fuzzy logic. Powers and McMullen (2000) use DEA as a decision-making tool to examine 185 large cap stocks with eight attributes to provide information on their relative efficiency scores, and target levels required by them to be efficient. Haslem and Scheraga (2003) use DEA to identify the efficient and inefficient large market cap mutual funds. They also identify the financial variables that differ significantly between efficient and inefficient funds, and determine the nature of these relationships. Their findings are consistent with characterizing the investment style of large cap DEA performance-efficient funds as predominantly value rather growth.
Wong et al. (1991) describe, from a general system-design perspective, ANN approach for a stock selection strategy. Fan and Palaniswami (2001) use the support vector machines (SVM) in a classification approach to ‘beat the market.’ Given the fundamental accounting and price information of stocks trading on the Australian Stock Exchange, they attempt to use SVM to identify stocks that are likely to outperform the market by having exceptional returns. The equally weighted portfolio formed by the stocks selected by SVM has a total return of 208 per cent over a five-year period, significantly outperformed the benchmark of 71 per cent.
Posted in :
