Use familiar terms for the options

[haziqj]

Change:

• delt to stop.crit

Decided to keep report.int instead of show.prog. The reason is that report.int is an option to determine how often the EM steps are reported. show.prog would behave to the the already available option silent. However, I've found that some R functions use quietly or quiet instead of silent. I'll leave this for now...

But will look at other options if some of these can be changed to a more meaningful name.

Although I think once the I-prior package is properly documented then there won't be any confusion.

[haziqj]

For convenience, here is a list of all available user options and their default values as of v1.3.2

• parsm=T. Option to model interactions parsimonious way or non-parsimonious way (set parsm=F). Obviously only works when interactions are present, and this option is ignored when there aren't any interactions present.
• one.lam=F. Option to set only a single scale parameter instead of one scale parameter for each covariate.
• maxit=50000. The maximum number of iterations the EM will run. The EM will stop either when the stopping criterion is met, or the maximum iterations is met.
• delt=0.001. The stopping criterion. This is the difference between successive log-likelihood values. _to be changed to stop.crit_.
• report.int=100. This sets how often the EM should report back to the user. By default, it is every 100 EM steps.
• silent=F. This lets the iprior() function run without any reporting, useful when using in a simulation scenario.