Programs Available

Below are details of code developped as part of my research. Currently this covers methods for changepoint detection; estimating recombination rates from population genetic data; algorithms for perfect simulation of non-neutral population genetic models; and a range of computational statistical algorithms.

Changepoint Detection

Fast algorithms for change detection via (optimally) minimising a penalised cost (or maximising a penalised likelihood) criteria:

  • The FPOP algorithm, which detects change-in-mean for univariate data under an L2 loss function.
  • The RPOP algorithm, which detects change-in-mean for univariate data in the presence of outliers.
  • The CPOP algorithm, which datects change-in-slope for univariate data.
  • Also see the methods available within the changepoint package on CRAN.

Changepoint algorithms for specific applications:

Population Genetics