Amandeep Singh markAmandeep Singh
Vol. 01 / Independent researchStatistics · Risk · Law · Institutions

Amandeep Singh / Researcher

Amandeep Singh

Data Scientist · Actuarial Analyst · Statistician

Mathematically driven problem-solver applying ML/AI to real-world decisions, with strength in statistical modeling, data interpretation, and building empirical, scalable solutions. Thrives in research-intensive, cross-functional teams and communicates complex ideas clearly.

Portrait of Amandeep Singh
Independent researcherVirginia, USA
Current inquiry

How uncertainty, institutional structure, and public evidence shape consequential decisions.

  • Supervised learning
  • Feature engineering
  • Model interpretability
  • Experiment design
01Current / Featured work

One argument, given the room it deserves.

The newest long-form essay from the research notebook, presented with its sources, method, and revision record.

Full archive
A sculpted posterior landscape with a chain of beads tracing Markov chain motion across multiple modes.
Essay 003Technical explainer

Markov Chain Monte Carlo: Intuition, Algorithms, Diagnostics, and Why the Chain Moves the Way It Does

A guide to MCMC with interactive visuals: why Bayesian inference becomes an integration problem, how Metropolis-Hastings works, and how geometry shapes mixing and diagnostics.

Published
April 15, 2026
Reading time
14 minutes
Topic
MCMC
02Research areas

Three departments. One connected body of work.

Essays and projects are organized by the question they investigate—not by format alone.

Browse dossiers
01Bayesian MethodsEssays and projects on Bayesian statistics, nonparametrics, inference, and technical modeling.03 records02Macro RiskWork on macroeconomic tail risk, actuarial design, insurance systems, annuities, and financial stress.02 records03India, Law & PoliticsEssays on Indian law, dissent, political structure, sovereignty, caste, and state power.00 records
03Selected research

Methods, evidence, results—and the limits around them.

Technical investigations developed across probability, finance, actuarial systems, and applied statistical science.

All projects
Springer Nature square cover image for New Trends in Bayesian Statistics, the volume containing Bayesian Nonparametric Estimation of Time-Varying Macroeconomic Tail Risk.

Project 01 / Featured

Bayesian Nonparametric Estimation of Time-Varying Macroeconomic Tail Risk

A Bayesian nonparametric framework for tracking how macroeconomic downside risk evolves over time, using a time-dependent Dirichlet process mixture on long-run cross-country consumption growth data.

Bayesian NonparametricsDirichlet Process MixturesMacroeconomic Tail Risk
04Writing / 003 essays

An editorial record of methods, risk, and institutions.

Chronological, searchable, and organized for reading rather than browsing a wall of cards.

Search the archive
  1. 002Bayesian vs. Frequentist: Two Ways of Thinking About UncertaintyTwo dominant frameworks of statistical inference—Frequentist and Bayesian—start from the same data but differ in what probability means and how uncertainty is quantified.4 min
  2. 001FIA and MYGA: General Account Economics, Crediting Design, and Actuarial Risk ManagementA technical deep-dive into fixed indexed annuities (FIAs) and multi-year guaranteed annuities (MYGAs): crediting mechanics, option budgets, hedging/ALM interactions, behavioral risk, and the reserve/capital constraints that determine what designs are actually feasible.9 min

Publication record / At a glance

03
Published essays
02
Research projects
03
Research domains
05 / Correspondence

For research, writing, and carefully framed problems.

Email works best. Include the question, relevant context, and the kind of exchange or collaboration you have in mind.

Start a conversation ↗