I am using the moments-based approach to calculate the risk premium of growing a specific crop. As proposed by Chavas (Kim and Chavas,2003 and Chavas et al,2009) I used CRRA risk preference. Then incorporate moments (variance and skewness) into the equation used by Antle (1987). I got a problem as variance and skewness are in large numbers (into the power of 10^7 sometimes) the calculated risk premium is a huge number. I guessed that there is something wrong here. Can anyone suggest of a source of information of the detailed risk calculation using the moments-based approach?