Optimizing Time Series Forecasting for AI Job Market Dynamics: A Comparative Study of SARIMA and Prophet with Cross-Validation

Abstract

The rapid growth of Artificial Intelligence (AI) has led to significant shifts in the job market, making accurate forecasting essential for workforce planning and policy-making. This study aims to compare the effectiveness of two time series forecasting models, SARIMA and Prophet, in predicting AI job market trends. The primary focus was to evaluate their performance using RMSE and Time Series Cross-Validation (TSCV). The SARIMA model outperformed Prophet, particularly when capturing seasonal patterns, achieving an RMSE of 83.42, compared to Prophet's 136.01. The results demonstrate that SARIMA’s parameter tuning and seasonality handling contribute to more reliable forecasts, while Prophet’s flexibility offers advantages in non-linear data scenarios. The study provides valuable insights for stakeholders in adapting workforce strategies, particularly in AI-driven sectors, by recommending SARIMA for seasonal trends and Prophet for volatile, non-linear data.