Skip to content

This repository evaluates the performance of volatility and Value-at-Risk (VaR) models in forecasting risk for two portfolios: an equally weighted portfolio of nine developed market currencies and a portfolio including the Argentinian peso as an emerging market currency.

Notifications You must be signed in to change notification settings

shreyasxi/Forecasting-and-Backtesting-Risk-Models-for-FX-Portfolios

Repository files navigation

Forecasting and Backtesting Risk Models for FX Portfolios

License: MIT

Author: Shreyas Urgunde
Date: February 2025
Institution: Warwick Business School


📌 Objective

This repository evaluates the effectiveness of volatility forecasting models (Moving Average, EWMA, GARCH) and Value-at-Risk (VaR) estimation methods (parametric and historical simulation) in capturing FX risk for two distinct portfolios:

  1. Dollar Portfolio: An equally weighted synthetic portfolio of nine developed market currencies.
  2. Emerging Market Portfolio: A portfolio incorporating the Argentine Peso (ARS) to analyze risk dynamics in volatile emerging markets.

The project aims to empirically assess how well traditional models adapt to stable (developed) and highly volatile (emerging) currency environments, providing insights for FX risk management.

🛠️ How to Use

Step 1. Clone the Repository

git clone https://github.com/shreyasxi/Asset-Pricing-in-the-Bond-Market.git
cd your-repo-name

Step 2. Prepare the Data

Data_file.xlsx contains the relevant data to run this analysis, just make sure to change the file path in your script.

Step 3: Run the Analysis in your system

Run the script to generate forecasts, VaR estimates, and backtesting results using RStudio or an R-supported IDE like Visual Studio Code.

Step 4: Customize or Extend

This framework can be adapted for further research or applications:

  • Stress Testing: Introduce extreme historical or hypothetical scenarios (e.g., currency crises, geopolitical shocks) to evaluate model robustness under tail events.
  • Alternative Distributions: Replace parametric VaR assumptions (e.g., Normal) with heavy-tailed distributions (Student’s t, Generalized Pareto) to better capture emerging market risk.
  • Regime-Switching Models: Account for structural breaks or volatility regimes in emerging markets using Markov-switching GARCH.

Feel free to modify the code or dataset to adapt the analysis for your own independent research or projects. You do not need to take my permission at all!

About

This repository evaluates the performance of volatility and Value-at-Risk (VaR) models in forecasting risk for two portfolios: an equally weighted portfolio of nine developed market currencies and a portfolio including the Argentinian peso as an emerging market currency.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages