<?xml version="1.0" encoding="utf-8" ?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:r="https://r-universe.dev"><channel><title>andrewwilson201.r-universe.dev</title><link>https://andrewwilson201.r-universe.dev</link><description>Recent package updates in andrewwilson201</description><generator>R-universe</generator><image><url>https://github.com/andrewwilson201.png</url><title>R packages by andrewwilson201</title><link>https://andrewwilson201.r-universe.dev</link></image><lastBuildDate>Fri, 27 Feb 2026 21:54:53 GMT</lastBuildDate><item><title>[andrewwilson201] actuary 0.1.2</title><author>andrewm.wilson@gmail.com (Andrew Wilson)</author><description>Provides actuarial modeling tools for Monte Carlo loss
simulations, loss reserving, and reinsurance layer loss
calculations. It enables users to generate stochastic loss
datasets with customisable frequency and severity
distributions, fit development patterns to claim triangles, and
calculate reinsurance losses for occurrence and aggregate
layers with user-defined retentions, limits, and
reinstatements. For development pattern selection, the package
includes a machine learning approach that evaluates multiple
reserving models using holdout validation to identify the
best-fitting pattern based on predictive accuracy, this is
based on the algorithm described in Richman, R and Balona, C
(2020)&lt;https://www.ssrn.com/abstract=3697256&gt;.</description><link>https://github.com/r-universe/andrewwilson201/actions/runs/28646622335</link><pubDate>Fri, 27 Feb 2026 21:54:53 GMT</pubDate><r:package>actuary</r:package><r:version>0.1.2</r:version><r:status>success</r:status><r:repository>https://andrewwilson201.r-universe.dev</r:repository><r:upstream>https://github.com/andrewwilson201/actuary</r:upstream></item></channel></rss>