Out-Of-Sample Time Series Forecasting: OOS introduces a comprehensive framework for time series forecasting with traditional econometric and modern machine learning techniques.

Forecast uncertainty based on model averaging

This repository contains the R-Package for a novel time series forecasting method designed to handle very large sets of predictive signals, many of which may be irrelevant or have only short-lived predictive power.

Official Python implementation of the Pioneer Detection Method (PDM) — convergence-based expert aggregation and opinion pooling under structural change. Code for Vansteenberghe (2026), The Geneva Papers 51(1).

This code mainly computes the forecast of headline inflation using different aproaches. Likewise presents the forecast evaluation for each model along different points in a span period.

End-to-End Python implementation of Shin (2026)'s evaluator-locked agentic loop for transparent empirical research. Combines LLM-driven specification search with immutable evaluation harnesses, penalized regression (peLASSO), and Diebold-Mariano testing on ECB forecast data. Addresses the "garden of forking paths" crisis in AI-driven economics.

Honours research project for Sapphire Li (2023)

End-to-End Python implementation of Liu & Cheng's (2026) methodology for U.S. Treasury yield curve forecasting. Combines Factor-Augmented Dynamic Nelson-Siegel models, High-Dimensional Random Forests, and Distributionally Robust Optimization (DRO) for risk-aware ensemble forecasting under ambiguity.

A Codera Hackathon Project: One-month-ahead USDZAR forecasts, 261 out-of-sample origins 2021-2025. Forecast combination beats an AR(1) benchmark by 0.42% in RMSE, with a power analysis, a pairwise Diebold-Mariano matrix and a test suite for the leakage guarantees.

📈 Forecast U.S. Treasury yield curves with a robust machine learning approach, enhancing accuracy and decision-making in finance.