Regressor - Based Optimization of Auriferous Worth Forecasting

Kavipriya P., Shanthi S.

Abstract: Accurately Forecasting the future value of auric is important for investors and financial analysts because auric plays a vital role in protecting against inflation and currency fluctuations. This research introduces a fresh method for forecasting auric prices by utilizing Machine Learning methods with a particular emphasis on the Random Forest algorithm. The model seeks to understand intricate patterns and connections that impact auric prices and by using auric cost data and pertinent economic indicators. To enhance performance, we optimize the regressor's parameters using advanced techniques such as grid search and Bayesian optimization. This ensures that the LSTM network adapts effectively to the volatile and dynamic nature of gold prices. The proposed model is evaluated against traditional statistical approaches and other machine learning techniques, showing superior performance in terms of prediction accuracy, mean absolute error (MAE), and root mean squared error (RMSE). Our findings demonstrate the potential of regressor - based LSTM optimization in providing reliable and precise forecasts for auriferous worth, offering significant value to stakeholders in the financial markets

Keywords: LSTM Networks, Forget gate, Input gate, Regressor, Datasets