Purpose-The current study aims to investigate the impacts of two behavioral biases, namely, loss aversion and overconfidence on the performance of US companies. First, the impact of loss aversion on the economic performance of companies was assessed. Second, the impact of overconfidence on market performance was discussed. Design/methodology/approach-This study used around 6,777 quarterly observations on the population of US-insured industrial and services companies over the 2006-2016 period. Ordinary least squares (OLS) regression in two panel data models were used to test the hypotheses formulated for the study. Findings-It was documented that the loss-aversion bias negatively affects the economic performance of companies and this is achieved for both sectors. In contrast, the findings suggest that overconfidence positively affects market performance of industrial firms but negatively affects market performance in service firms. Further robust evidence was found that overconfidence bias seems to be dominant, and hence, investors may tend to be more overconfident rather than more loss-averse. Originality/value-This research can be extended by focusing on the following question: What is the impact of the contradictory (positive and negative) effects of an investor's loss aversion and overconfidence on the US company performance in case of realization of a stock market crisis or stock market crash?

Purpose The purpose of this paper is to detect quantitatively the existence of anchoring bias among financial analysts on the Tunisian stock market. Both non-parametric and parametric methods are used. Design/methodology/approach Two studies have been conducted over the period 2010–2014. A first analysis is non-parametric, based on observations of the sign taking by the surprise of result announcement according to the evolution of earning per share (EPS). A second analysis uses simple and multiple linear regression methods to quantify the anchor bias. Findings Non-parametric results show that in the majority of cases, the earning per share variations are followed by unexpected earnings surprises of the same direction, which verify the hypothesis of an anchoring bias of financial analysts to the past benefits. Parametric results confirm these first findings by testing different psychological anchors’ variables. Financial analysts are found to remain anchored to the previous benefits and carry out insufficient adjustments following the announcement of the results by the companies. There is also a tendency for an over/under-reaction in changes in forecasts. Analysts’ behavior is asymmetrical depending on the sign of the forecast changes: an over-reaction for positive prediction changes and a negative reaction for negative prediction changes. Originality/value The evidence provided in this paper largely validates the assumptions derived from the behavioral theory particularly the lessons learned by Kaestner (2005) and Amir and Ganzach (1998). The authors conclude that financial analysts on the Tunisian stock market suffer from anchoring, optimism, over and under-reaction biases when announcing the earnings.

Keywords:Week of the year effect, Day of the week effect, Week of the month effect, Month of the year effect, Stock return index (TUNINDEX) ABSTRACTThe aim of this study is to investigate the presence of seasonal market anomalies (calendar anomalies) and to analyze their effects on the behavior of financial investors in terms of decisions and profit on the Tunisian market during the entire period that starts on January 2003 and ends on 31 December 2015. This work examines four calendar effects which are the new week of the year (WOY), the day of the week (DOW), the week of the month (WOM) and month of the year (MOY) effects using daily data of Tunisian Stock Market Index (TUNINDEX closing price) and dummy variables based on a GARCH (1,1) regression model adopted by Levy and Yagil (2012) to demonstrate whether the anomalies exist on the Tunisian market. The findings show that the returns for Friday are always positively significant. In contrast, the returns for October are almost negatively significant and low compared to other months. We also find that market calendar anomalies are clues to help investors improving their trading strategies and timing their investments to make abnormal profits. Contribution/ OriginalityIn this article, we examine the effect of financial market anomalies in the context of Tunisian financial market, specifically four calendar anomalies which are the week of the year effect, the month of the year effect, the week of the month effect and day of the week effect on the behavior of financial investors in terms of decisions and profit.

No abstract

No abstract