Yi-Nung,

Thank you for the prompt fix - all good now.

As an aside, how are you estimating the initial values in the HoltDES routine- a regression based on a sample of values? Thanks.

Best,

George