ºÚÁϳԹÏ×ÊÔ´

Biostatistics

I am interested in applications of statistics and probability to biology, especially to marine ecology. I have developed methods for line transect surveys and CKMR (close-kin mark recapture) methods in particular. Key reference for CKMR:

1. Bravington, MV, Skaug, HJ, Anderson, EC (2016). Close-kin Mark-Recapture. Statistical Science.

Computational statistics

My opinion for a long time has been that Statisticians should have Automatic Differentiation (AD) in their toolbox. AD is the same type of technology as "back-propagation" used in Deep Learning. My conjecture has been that a combination of AD and the Laplace approximation is a powerful platform for fitting models with continuous latent variables. Software projects in which this idea is implemented are:

• TMB:

• ADMB:

Key references for the AD-Laplace:

1. Skaug, H.J., Fournier D.A. (2006) Automatic Approximation of the Marginal Likelihood in non-Gaussian Hierarchical Models. Computational Statistics and Data Analysis.
2. Kristensen, K., Nielsen, A., Berg, C. W., Skaug, H., & Bell, B. (2016). TMB: automatic differentiation and Laplace approximation. Journal of Statistical Software.

Editorial services

co-Editor in chief of Scandinavian Journal of Statistics (2018–2021)

Short courses / workshops (last 5 years)

• CKMR (Close-kin mark recapture) course (co-thought) Dalhousie University, Halifax, July 2019.
• TMB (Template Model Builder) course (co-thought) at IMR Bergen, June 2019.
• Modelling in TMB (Template Model Builder). Course (co-thought) at IMR Bergen,  June 2018.
• An introduction to fish stock assessment in TMB (Template Model Builder). Course given Oct. 2017 in La Jolla

Teaching at ºÚÁϳԹÏ×ÊÔ´ (last 5 years)

• Spring 2018: STAT202 Biostatistics
• Fall     2017: STAT110 Introduction to statistics
• Spring 2017: STAT111 Statistical Methods
• Fall     2016: STAT110 Introduction to statistics
• Spring 2016: STAT202 BiostatisticsFall     
• 2015: STAT110 Introduction to statistics
• Spring 2015: STAT211 Time series

Google Scholar: 

From CRIStin (Current research information system in Norway):