We test whether a Genetic Algorithm (GA) can find profitable investment strategies based on prior stock returns and earnings surprises. We add to the argument whether momentum investing profits are a statistical illusion. The performance of the optimized momentum portfolios is evaluated before and after trading costs, during different time periods, over two market states, and after adjusting for risk. The GA optimization improves the annual returns of the momentum strategies by 2% to 6%. After considering transaction costs, both price and earnings momentum portfolios do not appear to generate abnormal returns. Positive risk-adjusted returns net of trading costs are documented solely in the �up� markets for a portfolio long in prior winners only.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados