Financial market volatility is driven by a complex interplay of factors, many of which exhibit nonlinear relationships with market behavior. Traditional linear models often struggle to adequately capture these intricate dynamics, particularly when analyzing heterogeneous data sources such as stock prices, interest rates, and macroeconomic indicators. In this paper, a nonlinear regression approach is proposed, specifically using generalized additive models, to better understand the impact of such diverse data on market volatility. By incorporating data from multiple sources, including historical stock prices and interest rates, significant nonlinear relationships between stock returns and interest rates with market volatility are revealed in this paper. The smooth terms for log returns show that volatility increases sharply during periods of negative returns, particularly during market downturns, which is consistent with the widely observed phenomenon of volatility clustering. By accounting for these nonlinear relationships, the model provides valuable insights into how varying economic conditions impact market stability, informing risk management strategies and policymaking.
