Portfolio Optimization Considering Behavioral Stocks with Return Scenario Generation

This study extends the application of behavioral portfolio optimization by estimating return stocks (B-stocks). With cause-and-effect relationships respective irrational behaviors on stock price movements and unique information provided B-stocks in terms knowing with a calculated probability when (time duration) specific effect (e.g., positive cumulative abnormal return) after certain trigger point (cause behavior) is spotted, regression analysis applied duration to have more accurate estimates. To fit framework optimization, scenarios used for are generated utilizing analysis, based which safety-first scenario-based mixed-integer program obtain optimal portfolios. also proposes two new types corresponding operational definitions herding ostrich-effect, along previously identified over-reaction, under-reaction, disposition-effect B-stocks. Back-test results show that portfolios profitable can significantly outperform benchmark market.