Publication

Abstract : In the modern era of data-driven decision-making, accurately forecasting employment trends is essential for policymakers and businesses. Traditional forecasting approaches often fail to incorporate real-time financial indicators, such as stock market performance, which frequently act as early signals of sector-wise hiring patterns. This study proposes a hybrid framework that integrates portfolio optimization with machine learning–based stock prediction to forecast sectoral employment trends. Using historical stock data from the S&P 100 (2015–2024) and government labor statistics, key financial indicators such as returns, volatility, and Sharpe ratio were extracted and correlated with employment variations using supervised learning techniques. XGBoost regression and Random Forest classification models were employed for stock forecasting, while portfolio construction was guided by Efficient Frontier theory via PyPortfolioOpt. Experimental results indicate a strong association between optimized portfolios and positive employment growth in the IT, Healthcare, and Consumer Goods sectors. The proposed integrated model facilitates real-time forecasting of labor demand using financial signals, offering strategic insights for investors, businesses, and workforce planning agencies.