Re: [Gretl-devel] Mann-Kendall test