Skip to contents

TODO: CHANGE THIS TO A USER SUMMARY OF THE PACKAGE

NOTE THAT FOR CONINTUOUS TIME logmu USES DATEY PACKAGE

The logmu package provides high-performance actuarial mortality experience analysis and model fitting, built on a flexible mortality framework incorporating time-based covariates and arbitrary proportional hazards models and with particular support for postcode-based socio-economic mortality models.

A/E analysis includes confidence intervals and residuals with visualisation. Model fitting features include covariate clustering, maximum likelihood fitting and model selection using AIC.

Weighting #' (e.g. amounts vs lives), probabilistic similarity weighting (e.g. down-weight older data) and optionally time-based inclusion criteria (e.g. sub-setting experience by age) are supported throughout.

Calculations use SIMD vectorisation and multi-threading for performance.

See also

  • TODO

  • vignette("XXX") for a worked introduction.

  • The series of articles starting with vignette("Measures matter") for the logmu theoretical framework.

Author

Maintainer: Tim Gordon tim.gordon@btinternet.com (ORCID)

Authors: