In the present paper, we estimate a monthly index of aggregate skewness preference. This index is derived from the representative agent's nonparametric probability weighting function. Consistent with findings from lab experiments, probability weighting functions depend on sentiment and their average is inverse-S shaped. Induced skewness preference varies over time and explains the size (value) premium over and above three (two) Fama and French factors and a momentum factor. Closed-end fund discounts signal a desire to underdiversify during times of high skewness preference. Higher skewness preference leads to more low quality IPOs with higher delisting probability.