Deep Learning vs. Statistical Models for Multi-Horizon Price Forecasting of Second-Hand Electronics: A Systematic Benchmark

Name: Deep Learning vs. Statistical Models for Multi-Horizon Price Forecasting of Second-Hand Electronics: A Systematic Benchmark
Uploaded: 2026-10-15T23:49:24.366074+00:00
Duration: 405 s
Description: This study benchmarks eleven models, including deep learning architectures like N-BEATS and statistical methods, to forecast used electronics prices over six time horizons up to 365 days. Findings show N-BEATS consistently outperforms other models beyond 30 days with the lowest error rate, proving deep learning is superior for capturing long-term trends in this market.

This study benchmarks eleven models, including deep learning architectures like N-BEATS and statistical methods, to forecast used electronics prices over six time horizons up to 365 days. Findings show N-BEATS consistently outperforms other models beyond 30 days with the lowest error rate, proving deep learning is superior for capturing long-term trends in this market.

Open the full episode