Zero-Shot Forecasting Mortality Rates: A Global Study

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Petnehazi, Gabor, Shaggah, Laith Al, Gall, Jozsef, Aradi, Bernadett
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910955053514752
author Petnehazi, Gabor
Shaggah, Laith Al
Gall, Jozsef
Aradi, Bernadett
author_facet Petnehazi, Gabor
Shaggah, Laith Al
Gall, Jozsef
Aradi, Bernadett
contents This study explores the potential of zero-shot time series forecasting, an innovative approach leveraging pre-trained foundation models, to forecast mortality rates without task-specific fine-tuning. We evaluate two state-of-the-art foundation models, TimesFM and CHRONOS, alongside traditional and machine learning-based methods across three forecasting horizons (5, 10, and 20 years) using data from 50 countries and 111 age groups. In our investigations, zero-shot models showed varying results: while CHRONOS delivered competitive shorter-term forecasts, outperforming traditional methods like ARIMA and the Lee-Carter model, TimesFM consistently underperformed. Fine-tuning CHRONOS on mortality data significantly improved long-term accuracy. A Random Forest model, trained on mortality data, achieved the best overall performance. These findings underscore the potential of zero-shot forecasting while highlighting the need for careful model selection and domain-specific adaptation.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13521
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Zero-Shot Forecasting Mortality Rates: A Global Study
Petnehazi, Gabor
Shaggah, Laith Al
Gall, Jozsef
Aradi, Bernadett
Machine Learning
Risk Management
Applications
This study explores the potential of zero-shot time series forecasting, an innovative approach leveraging pre-trained foundation models, to forecast mortality rates without task-specific fine-tuning. We evaluate two state-of-the-art foundation models, TimesFM and CHRONOS, alongside traditional and machine learning-based methods across three forecasting horizons (5, 10, and 20 years) using data from 50 countries and 111 age groups. In our investigations, zero-shot models showed varying results: while CHRONOS delivered competitive shorter-term forecasts, outperforming traditional methods like ARIMA and the Lee-Carter model, TimesFM consistently underperformed. Fine-tuning CHRONOS on mortality data significantly improved long-term accuracy. A Random Forest model, trained on mortality data, achieved the best overall performance. These findings underscore the potential of zero-shot forecasting while highlighting the need for careful model selection and domain-specific adaptation.
title Zero-Shot Forecasting Mortality Rates: A Global Study
topic Machine Learning
Risk Management
Applications
url https://arxiv.org/abs/2505.13521