Re: [Gretl-users] lo and mackinlay variance ratio test