Journal of the Royal Statistical Society: Series A

We are also in the butterfly site! (find us with the handle jrssa.bsky.social)

Discover the latest special issue from @JRSS_A: “Blockchains in Society,” showcasing research on blockchain analytics, digital assets, fraud detection, and cryptocurrency networks. @RoyalStatSoc Read the issue: oxford.ly/48gzHyB

New article with @albertocaimo!

Applied to Norwegian LFS data, the new estimator provides substantial precision gains for estimates of change over time in employment.

The method explicitly models autocorrelated sampling errors and time-varying wave biases, offering a robust alternative to state-space models and models all labour market categories simultaneously.

And in the spotlight, from Håvard Hungnes "A novel multivariate composite estimator for the Labour Force Survey" doi.org/10.1093/jrsssa…

🚶Hoffman & Block model emergent structures in human mobility — connecting individual movement decisions to macro-level patterns. #Mobility #NetworkScience → doi.org/10.1093/jrsssa… (10/10)

🚨 Lee et al. model crime on linear networks (roads, paths) using a spatio-temporal Dirichlet process mixture — because crime doesn't happen in open space. #CrimeMapping #BayesianNonparametrics → doi.org/10.1093/jrsssa… (9/10)

🧵 April (and late March) highlights from JRSS-A! New original articles spanning networks, crime, genomics, finance & more. Thread 👇 (1/10)

Applied to three-way mortality data, the proposed models yield coherent, nondivergent forecasts across subpopulations.

Authors address a key gap in mortality forecasting by modeling the multiway structure that standard approaches discard. The study develops low-rank tensor autoregressive models for parsimonious and interpretable mortality forecasting.

📉 Boonen & Chen bring low-rank tensor autoregression to mortality modelling — capturing age, time, and cause-of-death structure jointly. Major contribution for actuaries and demographers alike. #MortalityModelling #Actuarial doi.org/10.1093/jrsssa…

The authors: @d_ciganda @jakhmack. Institutional support: @MPIDRnews, @Udelaruy, @CEDemografia, @uni_tuebingen, @MPI_IS.

Estimated solely on aggregate data, our model successfully predicts out-of-sample micro outcomes like age at first sex and birth intervals

👶 Learning Individual Reproductive Behavior from Aggregate Fertility Rates via Neural Posterior Estimation Ciganda et al. use neural posterior estimation to recover individual reproductive behaviour from fertility rates. #Demography #NeuralNetworks doi.org/10.1093/jrsssa…