BSTS Package: Error in the Mean Absolute Percentage Estimate (MAPE) as Inf % in a Bayesian Inference with MCMC Plot using ggplot() in R

Hi Technocrat. I hope you are well! Could I please inquire what code you used and the values you inputted into the upper limit to achieve a low MAPE value. Thank you if you can help. Kind regards.

For sdy I was simply bracketing close to zero with the mean of the series (14.63) and seeing what direction the abscissa was going. I wouldn't have been lead to the point about log lambda by reading the docs, although drilling down does point to the PoissonPrior in the {Boom} package, a wrapper for the C++ library.

However, due to my delay getting back to you, I've lost the thread of our discussion. Start over on a new post with a reprex and mention me?

That takes as its argument the mean of the Poisson distribution expected with optional upper and lower bounds.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.