Kevin E. See and Elizabeth E. Holmes. 2015. Reducing bias and improving precision in species extinction forecasts. Ecological Applications 25:11571165.


R code for simulation implementation and parameter estimation.
Ecological Archives A025-069-S1.


File list (downloads)


Kevin E. See
Quantitative Consultants, Inc.
Northwest Fisheries Science Center
2725 Montlake Boulevard East
Seattle, WA 98112

Elizabeth E. Holmes
Conservation Biology
Northwest Fisheries Science Center
National Marine Fisheries Service
2725 Montlake Boulevard East
Seattle, WA 98112

File list

EstimationComparison.R (MD5: 5055e25ebc61fe8dbf012b48198eadda)

Simulations.R (MD5: 56eef6b34cb59a85d5072b0b20e645ec)

CreateFigures.R (MD5: 2c19a14d1ac5887ba544c2937d8639ca)


EstimationComparison.R - This code was used to simulate time-series of length 30 time-step and either 1, 2, or 3 replicated observations at each time-step. Parameters were then estimated using three different methodologies: restricted maximum likelihood (REML) based on first differences, REML based on second differences and maximum likelihood using the Kalman filter, as implemented in the MARSS R package.

Simulations.R - This code was used to simulate time-series of varying length, with varying numbers of replicated observations, under two different rates of decline and several different process and non-process error variances. Estimates of trend, process and non-process error variances were then made with the MARSS R package.

CreateFigures.R - This code was used to create the figures that appear in the manuscript and appendices.