Explaining Trade Volatility: A Hybrid Machine Learning and SHAP-Based Interpretability Analysis of India's Foreign Trade (1990–2025)

Balasubramanian S; Natarajan M

Explaining Trade Volatility: A Hybrid Machine Learning and SHAP-Based Interpretability Analysis of India's Foreign Trade (1990–2025)

Author(s)	Mr, Balasubramanian S, Dr. Natarajan M
Country	India
Abstract	Accurate trade forecasting and structural volatility attribution are vital for macroeconomic stability, tariff policy formulation, and central bank foreign reserve management. However, empirical trade modeling often struggles with trade-offs between linear interpretability and complex predictive capacity. This paper presents an empirical framework evaluating machine learning models alongside Explainable Artificial Intelligence (xAI) techniques to model monthly aggregate exports in India from April 1990 to March 2025 (420 observations). We construct a feature space comprising autoregressive lags, moving averages, rolling volatility metrics, trade balance ratios, and exogenous crisis indicators. Benchmark evaluations across Ridge Regression, Random Forest, Extra Trees, LightGBM, and XGBoost demonstrate that Ridge Regression achieves superior generalization on out-of-sample test data (2020–2025), yielding a Root Mean Squared Error (RMSE) of $3.6592 billion, a Mean Absolute Error (MAE) of $2.6889 billion, a Mean Absolute Percentage Error (MAPE) of 10.05%, and an R² score of 0.6278. Non-linear ensemble methods suffer from trend extrapolation bias due to non-stationary structural shifts across the 35-year horizon. Model attribution using Shapley Additive exPlanations (SHAP) reveals that medium-to-long-term trend indicators (Exports_MA_12, Imports_MA_12) and short-term momentum (Exports_Lag_2, Exports_Lag_6) dominate global predictions, while short-term rolling volatility indicators exert localized non-linear pressure during post-shock trade recovery phases.
Keywords	Foreign Trade Forecasting, Explainable Artificial Intelligence (xAI), SHAP Attribution, Ridge Regression, Econometric Feature Engineering, India Trade Volatility
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 7, Issue 6, November-December 2026
Published On	2026-11-04

View / Download PDF File

E-ISSN 3048-7641

doi

CrossRef DOI is assigned to each research paper published in our journal.

AIJFR DOI prefix is
10.63363/aijfr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 7 Isu 6 Cover Page Vol 7 Isu 4 Cover Page Vol 7 Isu 3

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About AIJFR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9898-79-39-59	editor@aijfr.com

Advanced International Journal for Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Explaining Trade Volatility: A Hybrid Machine Learning and SHAP-Based Interpretability Analysis of India's Foreign Trade (1990–2025)

Share this