Hybrid Markov weighted fuzzy kernel time series with red Piranha Walrus optimization for gold price forecasting

The price of gold is crucial to the world’s financial and economic systems; hence precise estimation of gold prices is essential. The current study proposes a hybrid Markov Weighted Fuzzy Kernel Time Series framework for gold price prediction, together with Red Piranha Walrus Optimization (MWFKTS-RPWO). Initially the input data is preprocessed and fed to the MWFKTS approach. It incorporates Markov models to capture temporal dependencies, fuzzy logic to handle uncertainty, and kernel methods to capture nonlinear relationships in gold price data. Additionally, RPWO is employed to optimize model parameters. The proposed MWFKTS-RPWO model demonstrates superior performance with a training time of 60–90 s, inference time of 1–2 ms per sample, and memory usage of 200 MB. Compared to existing methods, it offers an optimal balance between computational efficiency and accuracy. As a result, the proposed method is a superior choice for managing and forecasting gold prices.