نقش برجستگی در شکل گیری نمونه کارها

We analyze the likelihood of a stock being included in an investor's portfolio, utilizing a dataset which holds the information opportunity set constant for each of the over 1000 student investors in our sample. Investors rely on the availability heuristic: salience (the number of stories in the national press about a stock in the month before the portfolios are formed) captures over 50% of the variation in our dependent variable.

How do investors choose stocks for their portfolios? Why do they prefer some stocks over others? Standard finance theory tells us how investors should form their portfolios. Investors should maximize their utility by optimizing the trade-off between return and risk amongst assets in their investment opportunity set (Markowitz, 1991). Investors appear to ignore Markowitz's sage advice. The weight of evidence demonstrates that investors do not form their portfolios “optimally”. Investors do not diversify their portfolios; they hold too few stocks (De Bondt, 1998 and Barber and Odean, 2001). Even when their pension funds offer them a range of investments, investors tend not to take up these opportunities (Benartzi and Thaler, 2002). Investors choose what is familiar: they prefer to buy stocks that are geographically and culturally close to them (Grinblatt and Keloharju, 2001 and Coval and Moskowitz, 1999). Investors “put all their eggs in one basket” by investing in the companies they work for (Huberman, 2001). Investors trade too often, eating up potential profits in trading costs (Barber and Odean, 2000, Barber and Odean, 2001 and Barber and Odean, 2002). Even Markowitz is reported to have split his retirement fund between stocks and bonds to avoid regret in the future (Shefrin, 2000, p. 120). We utilize a unique dataset to provide further evidence of how investors choose stocks for their portfolios. The data allows us to hold the information opportunity set constant for each investor and, as such, provides a strong control for the analysis of the use of information in forming portfolios. We find strong support for the hypothesis that investors utilize the availability heuristic when selecting shares. We examine a dataset of 1412 portfolios of between five and ten stocks, chosen by investors between August 21 and August 24, 2003. The portfolios were selected by Australian university students in order to enter the BRW National Student Share Portfolio Award 2003.1 Each investor had the same goal: to maximize his or her absolute return over a twelve-month period.2 Each entrant paid $10 for each portfolio they entered in the game. Each entrant had an initial endowment of $200,000 from which they could invest in 5 to 10 fully paid shares or trust units (but not options, warrants or futures) listed on the Australian Stock Exchange. Each investor was allowed to enter up to five portfolios. No more than $40,000 could be invested in one stock. Additionally, no trading was allowed after the portfolio was “locked-in” on August 23, 2004. The winner3 – the student earning the greatest return between August 25, 2003 and August 20, 2004 – received $10,000 in cash. The investors in our sample might be expected to be relatively more sophisticated than the investing population at large: the sample is dominated by business majors, especially those majoring in Finance and Accounting. The breakdown of the sample by major may be found in Table 1. In general, the finance majors will have taken, or would have been taking, a course using a textbook such as Bodie, Kane and Marcus (2002). Many of the students identifying themselves as accounting majors would also be expected to have a double major with finance or, at a minimum, one finance course in their degree.4 Perhaps of most importance to our study, as we have noted previously, the information opportunity set is also the same for all investors. The variation in portfolios we observe will therefore capture the cognitive processes and abilities in utilizing the information opportunity set to meet the specific goal of maximizing terminal wealth.Our study focuses on investors' use of information. Evidence points to investors' decisions being influenced by information that is more readily recalled. Rather than balancing all potential information equally, investors have a propensity to rely on information that might more easily come to mind; such a cognitive process has been labeled the availability heuristic (Tversky and Kahneman, 1973). A useful analogy might be that of a student in a library grabbing the first books he finds to complete an assignment rather than conducting a thorough and thoughtful review of the literature. Such behavior confuses salience with substance. There is empirical support for the notion that investors' behavior is consistent with actions driven by the availability heuristic. Klibanoff, Lamont and Wizman (1998) found that the amount of space devoted to a country on the front page of the New York Times reduces the discount in closed-end-funds of assets from that country. Fehle, Tsyplakov and Zdorovtsov (2003) have found evidence of significant abnormal returns for firms following their advertising during the Super Bowl. More generally, Barber and Odean (2006) find that news about stocks generates higher abnormal trading volume but that more sophisticated investors are less likely to be influenced by news. Intriguingly, Shiller (1987) found that investors were thinking about the Crash of 1929 before the Crash of 1987; perhaps the increasing salience of a market break in the past played a role in the events of 1987? Boyd (2001) argues that investors also use a recognition heuristic, choosing “familiar” over “less-familiar” stocks5 that may be associated with superior returns in “bull” markets.6 A number of studies have provided empirical evidence that investors choose stocks that are familiar to them ( Grinblatt and Keloharju, 2001, Huberman, 2001 and Coval and Moskowitz, 1999). We find that the salience of a company, proxied by the number of stories in the national press about that company in the month before portfolios had to be chosen (July 24, 2003 to August 24, 2003), is the predominant influence in determining the likelihood of inclusion of a company in an investor's portfolio. The more stories there are about a company, the more likely it is to be chosen. As investors become more sophisticated, the role of salience, though still statistically significant, falls.7 Investors also have information about the previous returns of the companies they might invest in. Therefore, we analyze the influence of past returns (both in the immediate run-up to the time investors had to choose their portfolio and in the preceding year).8 Our analysis makes no claims about the role of momentum as a priced-factor in the cross-section of returns of Australian stocks.9 Rather, we take our lead from the growing body of evidence indicating that investors condition their behavior on past returns. For example, Grinblatt and Keloharju (2000) found that classes of investors appear to follow different trading strategies: in their sample of investors in the Finnish market, domestic investors, particularly households, were found to follow contrarian strategies while foreign investors tended to be momentum traders. The findings from Finland are supported by those of a recent US study which also found that individuals tend to be contrarian (Kaniel et al., 2004). In the following section, we describe our data and discuss our methodology. We present the results of our analysis in Section 3 of the paper. Section 4 concludes our work.

Our dataset of students entering the BRW National Student Share Portfolio Award 2003 provides a unique opportunity to study the investment decisions of a large number of investors. Given the majors of the students we observe, we expect them to be relatively well-informed about markets and investment strategies. Our study holds the information opportunity set constant for each investor and, as such, provides a strong control for the analysis of the use of information in forming portfolios. We believe that we have found strong evidence consistent with investors relying on the availability heuristic in forming their portfolios. We find that the more news items there are about the stock, the more likely it is to be included in an investor's portfolio. Indeed, over 50% of the variation in the likelihood of a stock being included in an investors portfolio is explained by our proxy for the availability of information, salience. As investors become “savvier”, their propensity to utilize the availability heuristic in the cognitive process of forming portfolios diminishes. The explanatory power of salience in capturing the likelihood of being included in an investor's portfolio is lower, as measured by the value of the adjusted R2, for our sample of streetwise investors when compared to the ingénues in our sample. We also considered a measure of abnormal salience and found that was insignificant in our analyses; we therefore concluded that it is the level of attention that determines investors' reliance on the availability heuristic, rather than the change in coverage, that grabs investors' attention.