Niraj Neupane

Author · Wall Street Feed

Niraj Neupane

Chartered Accountant (ICAI) · Quantitative Finance Researcher · AI/ML in Financial Markets

  • Quantitative Researcher, Korvane
  • Financial Economist, Calderyn Institute
  • Founder, Wall Street Feed
  • Contributing Author, The Financial Gurkha
  • Business Correspondent, Times Asian
  • Quantitative Finance
  • Financial Econometrics
  • AI/ML
  • Trading
  • Risk Management
  • Model Validation
  • Financial Technology

About

Niraj Neupane is a Chartered Accountant (ICAI) and quantitative finance researcher working at the intersection of financial economics, financial econometrics, artificial intelligence, machine learning, quantitative modeling and financial markets. He is a Quantitative Researcher at Korvane, a Financial Economist at Calderyn Institute and the founder of Wall Street Feed. He is also a Contributing Author at The Financial Gurkha and a Business Correspondent for Times Asian, writing on markets, risk and the global economy.

His research examines how quantitative and AI/ML methods can be developed, tested, validated and responsibly applied across financial-market activities, including quantitative research, systematic trading, market and liquidity risk management, financial forecasting, regulatory and compliance applications, and financial-model governance.

A central theme of his work is the reliability of quantitative and machine-learning models: whether models that perform well under historical or normal conditions continue to provide useful, robust, interpretable and defensible results when markets experience volatility, liquidity changes, structural shifts and stress.

Background

Niraj combines a Chartered Accountant’s grounding in financial reporting and audit with academic training in economics, finance, financial economics and quantitative methods. His work brings together financial econometrics, statistical modeling, machine learning, financial engineering, programming and practical financial-market analysis.

A core area of his work is the development and validation of quantitative models: how models perform across different market regimes, how model selection can introduce overfitting, how forecasts become miscalibrated, and how validation frameworks can reveal weaknesses that conventional performance measures miss.

His broader agenda asks how artificial intelligence, machine learning and automation can be integrated into financial markets while maintaining high standards of reliability, transparency, validation, governance and human oversight.

Research & Working Papers

SSRN Working Paper · Abstract 7652318

When Backtests Lie: A Governance-Grade Validation Framework for Machine-Learning Value-at-Risk Models under Regime Shifts

A governance-oriented validation framework for machine-learning Value-at-Risk models. It extends conventional backtesting with regime-conditional coverage analysis, the Probability of Backtest Overfitting and conformal recalibration, to test whether aggregate evaluation can hide weaknesses in stressed or changing markets.

Read on SSRN →

SSRN Working Paper · Abstract 7534899

Systemic Tail Episode Risk: Persistence and Cumulative Severity of Synchronized Extreme-Loss Episodes Across Markets

Extends tail-episode analysis across markets to measure how often extreme losses arrive together, how long synchronized stress persists, and how severe it becomes in cumulative terms, a dimension that standard portfolio VaR can understate.

Read on SSRN →

SSRN Working Paper · Abstract 7502499

Conditional Tail Episode Risk: A Path-Dependent Framework for Extreme-Loss Episodes Beyond Value-at-Risk and Expected Shortfall

A path-dependent approach to tail risk that treats extreme losses as episodes, measuring their persistence and cumulative severity rather than single-day breaches alone.

Read on SSRN →

SSRN Working Paper · Abstract 7222958

ML-LiqVaR: A Liquidity- and Regime-Aware, Extreme-Value-Calibrated Gradient-Boosting Value-at-Risk Model for US Equities

A framework for one-day-ahead Value-at-Risk forecasting that combines volatility modeling, Extreme Value Theory, gradient-boosted quantile regression, liquidity and volatility features, and regime-aware modeling, while keeping rigorous statistical validation.

Read on SSRN →

SSRN Working Paper · Abstract 7170418 · July 2026

Backtesting Value-at-Risk Models Under SR 11-7: A Comparative Analysis of Kupiec, Christoffersen, and Basel Traffic-Light Tests Applied to S&P 500 Returns

A walk-forward empirical study of Value-at-Risk models on S&P 500 daily returns, comparing unconditional coverage, conditional coverage and Basel traffic-light backtests, and their relevance to model validation and risk management.

Read on SSRN →

Research Trajectory

01

ML-LiqVaR

Machine-learning methods for liquidity-aware, regime-aware and tail-sensitive Value-at-Risk.

02

Backtesting Under SR 11-7

Conventional statistical backtesting and model-validation approaches for risk models.

03

Tail Episode Risk

CTER and STER: persistence and cumulative severity of extreme-loss episodes, within and across markets.

04

Governance-Grade ML Validation

Regime robustness, model-selection risk, calibration and governance of machine-learning risk models.

Areas of Research

AI & ML in Financial Markets

Forecasting, quantitative research, risk measurement, anomaly detection and model development.

Quantitative Finance

Modeling, financial engineering, derivatives, portfolio analytics, systematic strategies and computational finance.

Financial Econometrics

Time-series analysis, GARCH-family volatility models, quantile regression, regime identification and inference.

Market & Liquidity Risk

Value-at-Risk, Expected Shortfall, liquidity-adjusted risk, tail risk and stress testing.

Quantitative Trading & Research

Systematic research, feature engineering, signal development, backtesting and strategy evaluation.

Model Validation

Coverage testing, benchmarking, calibration, stability and robustness analysis, and model-selection risk.

Regulatory & Compliance Analytics

AI/ML for surveillance, anomaly detection, regulatory analytics and compliance monitoring.

Responsible AI & Governance

Explainability, monitoring, documentation, auditability and responsible deployment in financial institutions.

Research Questions

  • How can AI and machine learning improve quantitative financial modeling while keeping high standards of reliability and validation?
  • How should machine-learning models be validated when markets go through structural change, volatility shifts, liquidity deterioration or stress?
  • Can conventional backtesting identify the weaknesses of flexible machine-learning risk models?
  • How can liquidity, volatility and market-regime information be built into financial risk models?
  • How can models be monitored for instability, overfitting, calibration drift and model drift?
  • How can explainability, validation and governance support the responsible use of AI in financial institutions, with appropriate human oversight?

Research Philosophy

A financial model should not be judged only by how well it performs under normal conditions.

It should also be evaluated for stability, calibration, robustness, interpretability and failure behavior when markets change. The objective is not simply to build models that perform well, but to understand when they work, when they fail, why they fail, and how those failure modes can be identified and managed.

Technical & Quantitative Methods

Financial Econometrics

Time-series analysisVolatility modelingGARCH-family modelsQuantile regressionForecastingRegime analysisStatistical testing

Machine Learning

Supervised learningGradient boostingQuantile MLFeature engineeringModel evaluationAnomaly detectionModel monitoring

Financial Engineering

Value-at-RiskExpected ShortfallDerivativesPortfolio analyticsStress testingTail-risk measurementLiquidity risk

Quantitative Research

Systematic researchBacktestingSignal developmentPortfolio constructionPerformance evaluationFinancial data analysis

Model Validation

BacktestingBenchmarkingCalibrationStability testingRobustness analysisOverfitting assessmentModel governance

Computational Finance

PythonC++RSQLMATLABPandasNumPyScikit-learnTensorFlow

Ventures

Quantitative Researcher

Korvane

A research and technology venture building AI-driven trade, risk and model-validation applications, spanning market-risk analytics, liquidity and stress analysis, monitoring, explainability and decision support.

Financial Economist

Calderyn Institute

A research and education initiative developing quantitative and AI/ML capability for financial markets: quantitative trading and research, financial risk, model validation, financial engineering and responsible AI.

Founder

Wall Street Feed

An independent publication covering Wall Street and global markets, pairing daily market news with research-driven analysis of risk, rates and the AI economy.

Writing & Media

Founder & Editor

Wall Street Feed

Daily coverage of Wall Street and global markets, with research-driven analysis of risk, rates and the AI economy.

Contributing Author

The Financial Gurkha

Long-form quantitative analysis of markets, tail risk and model validation, written for investors and risk professionals.

Business Correspondent

Times Asian

Business and economic reporting on markets, finance and the economies of Asia and the developing world.

Future Research Directions

Machine-learning risk measurementLiquidity-adjusted and regime-aware risk modelsExpected Shortfall backtestingAI-based stress testingExplainable AI for financial modelsML model validationQuantitative trading and systematic researchAI-assisted market surveillanceRegulatory and compliance analyticsModel monitoring and drift detectionResponsible AI and model governance

Articles on Wall Street Feed

About this research. Niraj Neupane contributes independent research on quantitative finance, financial economics, financial econometrics, artificial intelligence, machine learning, financial risk management, quantitative trading and financial-market technology. Research presented here may include working papers, preprints and research in progress. Unless explicitly identified as peer-reviewed or formally published, a paper should be treated as a working paper that has not undergone formal journal peer review. Findings are updated as methods, data and validation procedures evolve.

The research represents the author’s independent views unless a specific institutional or collaborative affiliation is explicitly identified, and nothing on this page is investment advice. Our sourcing, disclosure and corrections policies are set out in the Editorial Policy and Corrections Policy.