This paper integrates dynamic loss aversion and individual narrow framing in the investor utility function. The paper investigates investor behavior following the public announcement of earnings. To do this, the author optimizes the objective function of investor preference and analyzes to what extent the preference function generates a differential of value premium between stocks after the announcement of the firms' profits. Several simulations are used according to the different values considered for the preference parameters. The results show that in the short run, negative information infiltrates financial markets slowly as the adjustment coefficient of investor preference is below one. An important momentum premium is then noted. In the long run, reversal trends in returns generated by the preference framework confirm that investors feel very upset after the release of successive earnings of negative sign. In fact, losses that come after other losses are very painful for them as they have a dynamic loss aversion function.

Traditional finance literature often assumes that investors on financial market are rational. Hirshleifer (2001), Barberis and Thaler (2003) and Shiller (2002) argue that a `rational' investor is an investor who makes a relevant expectation based on all available information. This is done by updating beliefs using Bayes' rule. This investor must also make decisions according to the expected utility function of Von Newman and Morgenstern (1947), a utility function defined over wealth or consumption. However, experimental evidence suggests that investors, in a context of uncertainty, proceed to judgments which give rise to subjective probabilities (Kahneman and Tversky, 1974). In fact, after the release of new information on financial markets, investors commit cognitive errors when updating their beliefs of future profits. These errors are heuristics and biases which correspond to investor `cognitive rationality'. Moreover, Allais (1953), Ellsberg (1961) and Tversky (1969) find that investors have a different preferencefunction from that of the expected utility.

To explain these findings, a relevant literature argues that investor preference takes a form of a non-expected utility function.Examples of non-expected utility theories are the weighted utility theory of Chew and Mac Crimmon (1979) and Chew (1989), the theory of implicit expected utility of Chew (1989), the theory of regret aversion of Bell (1982), the theory of disappointment aversion of Gul (1991) and the prospect theory of Kahneman and Tversky (1979) and Tversky and Kahneman (1992).

Of all these non-utility theories, the prospect theory of Kahneman and Tversky (1979) is most emphasized in the finance literature. Indeed, prospect theory offers a descriptive framework to understand how individuals make decisions in the context of risk and uncertainty. It describes several statements that can influence the decisions of the investor. The main element of this theory is loss aversion. Loss aversion is a form of a first risk aversion which reflects the fact that people express more sensitiveness to losses than to gains of the same magnitude.

Behavioral Finance Journal, Risk Aversion Approach, Traditional Finance Literature, Financial Markets, Loss Aversion, Risk Aversion, Mental Accounting, Investment Constraints, Free Risk Assets, Dynamic Loss Aversions, Stock Markets, Market Equity, Asset Markets.