llanalysis: Functions for analyzing data contributions to the likelihood

In rplzzz/CovMitigation: COVID-19 infection model for Virginia

Description

These functions all take the table produced by the waicmode of gen_likelihood and summarize it by different groupings. This allows us to see whether certain counties or certain time periods are having an outsized contribution to the overall likelihood.

Usage

ll_bycounty(lldetail)

ll_bydate(lldetail)

ll_bydaterange(lldetail, n = 5)

cmpll_bycounty(lldetail, lldetail2)

cmpll_bydaterange(lldetail, lldetail2, n = 5)

Arguments

lldetail	Data frame of detailed log-likelihood contributions.
n	Number of ranges to divide dates into
lldetail2	Data frame of detailed log-likelihood contributions for a comparison scenario.

Details

To use these functions, run the parameters through a likelihood function generated by gen_likelihood with waicmode = TRUE. Then run the parameters you wish to investigate through that likelihood function and pass the result as lldetail.

Functions

• ll_bycounty: Report log-likelihood contributions by county

• ll_bydate: Report log-likelihood contributions by date

• ll_bydaterange: Report log-likelihood contributions by date ranges

• cmpll_bycounty: Compare log-likelihood contributions by county for two scenarios.

• cmpll_bydaterange: Compare log-likelihood contributions by date range for two scenarios

rplzzz/CovMitigation documentation built on June 7, 2021, 8:48 a.m.