Jere Koskela

About me

I'm a postdoc with at the Technische Universität Berlin, working on applications and inference of multiple merger coalescents in population genetics.

I finished my PhD in 2016 under the supervision of and . My thesis is .

Publications and preprints

J. Koskela, Multi-locus data distinguishes between population growth and multiple merger coalescents. Preprint
J. Koskela, D. Spanò and P. A. Jenkins, Consistency of Bayesian nonparametric inference for discretely observed jump diffusions. Preprint
J. Koskela, P. A. Jenkins and D. Spanò, Bayesian nonparametric inference for $\Lambda$ -coalescents: posterior consistency and a parametric method. To appear in Bernoulli
J. Koskela, D. Spanò and P. A. Jenkins, Inference and rare event simulation for stopped Markov processes via reverse-time sequential Monte Carlo. To appear in Statistics and Computing
J. Koskela, P. A. Jenkins and D. Spanò, Computational inference beyond Kingman's coalescent. Journal of Applied Probability 52.2 (2015), p. 519-537

Research grants

Co-PI, DFG SPP1819 Start-up grant, . Joint with (PI) and (Co-PI).

Talks

Consistency results for Bayesian nonparametric inference from processes with jumps, Probability and statistics seminar, TU Delft, 23 May 2017
Constructing a generalised coalescent process for joint modeling of high fecundity and natural selection, Oberseminar biologische Modelle und statistische Mechanik, TU Berlin, 15 May 2017
Distinguishing population growth from a skewed family size distribution in coalescent models, Oberseminar biologische Modelle und statistische Mechanik, TU Berlin, 28 November 2016
Bayesian nonparametric inference for $\Lambda$ -coalescents, Oberseminar Stochastik, Johannes Gutenberg Universität Mainz, 15 November 2016
Nonparametric Bayesian inference for $\Lambda$ -coalescents, Oberseminar biologische Modelle und statistische Mechanik, TU Berlin, 26 September 2016
Lossless Bayesian inference in infinite dimension without discretisation or truncation: a case study in $\Lambda$ -coalescents, International conference on Monte Carlo techniques, Paris, France, 8 July 2016
Bayesian inference for $\Lambda$ -coalescents: failures and fixes, SIAM UQ, Lausanne, Switzerland, 8 April 2016
Implementable Bayesian nonparametric inference via quotient spaces, Young Researchers' Meeting, University of 糖心TV, 23 February 2016
Efficient sequential Monte Carlo sampling of rare trajectories in reverse time, Algorithms & Computationally Intensive Inference seminar, University of 糖心TV, 12 February 2016
Pseudo-likelihood accelerated pseudo-marginal Metropolis-Hastings, Bayes Comp at MCMSki 5, Lenzerheide, Switzerland, 7 January 2016
Importance sampling rare trajectories in reverse time, EQUIP seminar, University of 糖心TV, 19 October 2015
Bayesian nonparametric consistency for jump diffusions, MASDOC Retreat, St Braivels Castle, Gloucestershire, 18 May 2015
Consistency of Bayesian nonparametric inference for jump diffusions, 4th Annual MASDOC-CCA Conference, University of 糖心TV, 16 April 2015
Measure-valued jump-diffusion sampling distributions via duality, MASDOC Retreat, Coalbrookdale, 19 May 2014
Time-reversal for efficient importance sampling of rare events in coalescent models, 3rd Annual CCA-MASDOC Student Conference, University of Cambridge, 18 March 2014
Importance sampling for the Spatial $\Lambda$ -Fleming-Viot process, Informal Workshop with Alison Etheridge, Oxford University, 20 June 2013
Importance sampling and $\Lambda$ -coalescents, Algorithms & Computationally Intensive Inference seminar, University of 糖心TV, 8 March 2013

Poster presentations

CRiSM workshop on Estimating Constants, University of 糖心TV, Coventry, UK, 20-22 April 2016
52nd Gregynog Statistical Conference, Gregynog Hall, Wales, 15-17 April 2016
Computational Statistics and Molecular Simulation, Cordeliers Convent, Paris, France, 2-5 February 2016
Models and Inference in Population Genetics, Department of Statistics, University of 糖心TV, UK, 14-16 September 2015
Annual Meeting of the Society for Molecular Biology and Evolution, Hofburg Palace, Vienna, Austria, 12-16 July 2015
Dynstoch, Centre for Research in Statistical Methodology, University of 糖心TV, UK, 10-12 September 2014
Biological sequence analysis and probabilistic models, Merton College, Oxford, UK, 16-19 July 2014

Teaching

At TU Berlin I have been a teaching assistant for two student seminars:

, 2017
, 2017

During my PhD I was a teaching assistant for the following modules:

(2013/14, 2014/15 and 2015/16)
(2013/14)
(2013/14 and 2014/15)
(2015/16)
(2013/14 and 2014/15)
(2013/14, 2014/15 and 2015/16)

Contact

koskela@math.tu-berlin.de

At the Brandenburg Gate