Ex. 1.4 Consider the daily log returns of American Express stock from January 1999 to December 2008 as in Exercise 1.1. Use the 5% significance level to perform the following tests: (a) Test the null hypothesis that the skewness measure of the returns is zero. (b) Test the null hypothesis that the excess kurtosis of the returns is zero.

the solution of book that i find on the coursehero:

: Daily log returns of American Express stock from 1999 to 2008. Skewness: test-statistic is t = 0.336/ p 6/2515 = 6.888 with p value 5.66 1012. Thus, we reject the null hypothesis of no skewness at the 5% level. Kurtosis: test-statistic is t = 6.486 24/2515 = 66.40, which is large and has a p value close to zero. Thus, we reject the null hypothesis of zero excess kurtosis. That is, the distribution of the log returns has heavy tails.

But i learned from book that can use this: > s3= skewness(da3$axp) > t= s3/sqrt(6/2515) > t [1] -6.888016 attr(,"method") [1] "moment" > pv=2*(1- pnorm(t)) > pv [1] 2 attr(,"method") [1] "moment

and pv is different