Effects of omitting information variables on optimal hedge ratio estimation: A note