Learning from the Storm: A Multivariate Machine Learning Approach to Predicting Hurricane-Induced Economic Losses

TL;DR

This paper introduces a comprehensive machine learning framework that integrates hurricane, environmental, and socioeconomic data at a fine spatial scale to predict economic losses from hurricanes, aiding disaster management.

Contribution

It develops a unified, multi-source data integration approach for predicting hurricane-induced economic losses using machine learning at the ZIP code level.

Findings

Accurately predicts economic losses using insurance claims as damage indicators.

Identifies key contributing factors among hurricane, environmental, and socioeconomic variables.

Provides insights into the relative importance of different factors for disaster mitigation.

Abstract

Florida is particularly vulnerable to hurricanes, which frequently cause substantial economic losses. While prior studies have explored specific contributors to hurricane-induced damage, few have developed a unified framework capable of integrating a broader range of influencing factors to comprehensively assess the sources of economic loss. In this study, we propose a comprehensive modeling framework that categorizes contributing factors into three key components: (1) hurricane characteristics, (2) water-related environmental factors, and (3) socioeconomic factors of affected areas. By integrating multi-source data and aggregating all variables at the finer spatial granularity of the ZIP Code Tabulation Area (ZCTA) level, we employ machine learning models to predict economic loss, using insurance claims as indicators of incurred damage. Beyond accurate loss prediction, our approach facilitates a systematic assessment of the relative importance of each component, providing practical guidance for disaster mitigation, risk assessment, and the development of adaptive urban strategies in coastal and storm-exposed areas. Our code is now available at: https://github.com/LabRAI/Hurricane-Induced-Economic-Loss-Prediction