Javier Meseguer
PhD Economist
Social Security Administration · American University
Last updated September 2, 2026
I am an economist working at the intersection of Bayesian statistics, time-series econometrics, and applied demography, with a focus on disability and retirement policy. This site collects my published papers; a pair of continually developed knowledge wikis that reflect my research focus over a 21+ year career at the Social Security Administration; and a large body of hands-on model implementations in Python and R that put those interests into practice — including replications of seminal articles across the relevant literatures.
The implementations have grown into three connected arcs. The first is Bayesian and econometric modelling — MCMC written from scratch alongside the standard engines, across regression, limited dependent variables, time series, spatial models and Bayesian nonparametrics. The second is machine learning, from trees and regularisation through neural networks to model evaluation, ending in operations research, where a forecast is judged by the decision it supports rather than by its error metric. The third is causal inference, where the question shifts from how well a model predicts to whether an effect can be identified at all.
Research Interests
- Bayesian Computational Methods and Inference
- Time series modeling of macroeconomic and financial variables
- Demographic forecasting
- Disability and retirement policy
- Labor economics
- Machine learning and neural networks
- Artificial intelligence and large language models (LLMs)