Asset ownership, investor characteristics, and the endowment effect

Abstract

We survey individuals regarding personal attributes and specific assets owned to identify characteristics associated with the endowment effect. Using eight separate models including multiple regression analysis and statistical modeling techniques including adjusted R-square, root-mean-square error (RMSE), Akaike’s Information Criterion (AIC), forward selection, backward elimination, and stepwise regression methods, we find that individuals who gamble more, have less formal training in investments, who are male, are less optimistic, have held the asset in question for longer periods of time, have a component of their income as commission based, and have a longer time period until retirement are more likely to exhibit a greater endowment effect bias.