How to arrange the output format of `summary.mcSimulation()` and `summary.welfareDecisionAnalysis()`

[l-goehring]

Dear @eikeluedeling,

I have the following issues emerging from #4:

1. Method summary.mcSimulation() calculates and returns $chance_loss, $chance_zero, $chance_gain (which are elements of $summary):
   • Are they really necessary and shall they still be kept? - If, yes
     • Semantically, they are part of the economic sphere. Thus, I would move them from summary.mcSimulation() to summary.welfareDecisionAnalysis(). Do you agree?
     • The current names are not consistent with the terminology used in the rest of the package. Thus, I would rename them to pLoss, pZero, pBenefit, respectively. Do you agree?
2. Default quantiles for methods summary.mcSimulation() and summary.welfareDecisionAnalysis():
   • Currently the default for summary.mcSimulation() is probs=c(0, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 1) But, the standard input variables are only described by the 90%-ci and optionally by the median (and the distribution type). This suggest to set the default of the quantiles to probs=c(0.05,0.5,0.95) and add information on the first three moments, i.e. mean, standard deviation and skewness (and maybe kurtosis) to describe the shape of the distribution intuitively. What do you think?

Best, Lutz

[eikeluedeling]

1. let's keep these measures in the package, but yes, please move them to summary.welfareDecisionAnalysis. Agree with the proposed name change.
2. I think we should keep the full range of quantiles. Doesn't hurt, and who doesn't want them all can use only some of them. But please add information on mean, sd and skewness. kurtosis also good. Good idea.

[l-goehring]

Thanx! - Will do so. - Lutz

[l-goehring]

Done with a small adaptation: $chance_loss and $chance_gain are not renamed to pLoss and pBenefit but to plPa, and pnbPa, respectively, because this is more consistent with the used names enbPa and elPa.