Paul Labonne

Research Associate, Department of Computing, Imperial College London

Time series predictive modelling & Software development for machine learning
[CV] [Github]

Working papers

• Risky news and credit market sentiment
  with Leif Anders Thorsrud
  [working paper]

• Flexible Negative Binomial Mixtures for Credible Mode Inference in Heterogeneous Count Data from Finance, Economics and Bioinformatics
  with Jamie Cross, Lennart Hoogerheide and Herman K. van Dijk
  [working paper]

Publications

• BayesMultiMode : Bayesian mode inference in R
  with Nalan Basturk, Jamie Cross, Peter de Knijff, Lennart Hoogerheide and Herman K. van Dijk
  Journal of Statistical Software, provisionally accepted
  [working paper] [R package]

• Asymmetric uncertainty: nowcasting using skewness in real-time data
  International Journal of Forecasting, 2024
  [paper] [code] [Python/JAX implementation]

• Bayesian mode inference for discrete distributions in economics and finance
  with Jamie Cross, Lennart Hoogerheide and Herman K. van Dijk
  Economics Letters, 2024
  [paper] [code]

• Temporal disaggregation of overlapping noisy quarterly data: estimation of monthly output from UK value-added tax data
  with Martin Weale
  Journal of the Royal Statistical Society (Series A), 2020
  [paper] [code]

• The financial foundations of the productivity puzzle
  with Jagjit S. Chadha and Amit Kara
  National Institute Economic Review, 2017

Software packages

• BayesMultiMode : Bayesian mode inference in R
   

Machine learning experience

Forecasting / Nowcasting / Big data (PCA, Ridge / LASSO) / Filtering (Kalman & Score driven) / Risk & Uncertainty / Bootstrap, Monte Carlo / Maximum Likelihood / Bayesian analysis / Backpropagation with automatic differentiation / NLP (word2vec) / Trend and seasonality / Temporal disaggregation