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

2026-10-07

AI Summary

This study addresses the absence of systematic benchmarks and the challenges posed by high volatility and data sparsity in price forecasting for second-hand electronics. We construct the first multi-horizon forecasting benchmark using Polish market data and propose a three-dimensional evaluation protocol to systematically assess ARIMA, LSTM, TCN, N-BEATS, and TFT models. Results demonstrate that a single N-BEATS model generalizes across all short-term horizons without requiring independent modeling. For 365-day forecasts, it achieves a MAPE of 8.51%, reducing the error by 43% compared to the best statistical baseline. This work establishes a new paradigm for time series forecasting in high-noise scenarios.

Abstract

Forecasting resale prices of used electronics is critical for subscription-based platforms where pricing errors translate directly into risk. Unlike structured financial markets, second-hand electronics exhibit high volatility, sparse listing histories, and non-normal price dynamics - yet no systematic time-series benchmark exists for this domain. This paper presents the first multi-horizon benchmark of statistical and deep learning forecasting models for used electronics price prediction. We use a large-scale dataset of daily price listings from Polish online marketplaces (January 2022 to March 2025, 100+ smartphone and laptop models) and evaluate eleven models across six horizons from 1 to 365 days, covering classical methods (ARIMA, ETS, Theta), recurrent and convolutional networks (LSTM, TCN), and modern deep architectures (N-BEATS, N-HiTS, TFT, PatchTST, Informer). Three complementary evaluation protocols assess trajectory fitness, one-shot endpoint accuracy, and cross-horizon transfer. N-BEATS achieves the lowest MAPE beyond 30 days, reaching 8.51% at 365 days versus 14.94% for the best statistical baseline - a 43% reduction. At short horizons (1-7 days), all models converge near 0.72% MAPE and the naive baseline remains competitive. A single N-BEATS model trained at 365 days generalizes to all shorter horizons, eliminating the need for horizon-specific models. N-BEATS and N-HiTS also demonstrate superior hyperparameter stability.

Problem

• Second-hand electronics
• Price forecasting
• Multi-horizon prediction
• Benchmark
• Time-series

Innovation

• Multi-horizon forecasting
• Second-hand electronics pricing
• Systematic benchmark
• N-BEATS
• Cross-horizon generalization