Application of Data Science on Stock Forecasting and Evaluation Using Artificial Neural Network

Stock exchanges are regarded as significant players in the financial sector in many countries. Most stockbrokers that execute trades use technical, fundamental, or time series analysis to try and predict stock values in order to advise clients. However, these strategies usually don't guarantee large profits because they go by trends rather than the most likely price. Thus, it is imperative to look into more sophisticated prediction methods. Although stock market prediction is a common utilization of artificial neural networks (ANN), stock market forecasting remains challenging due to the multitude of known and unknown factors that impact it. In this work, we proposed an ANN-based method for daily stock price forecasting. The results of statistical forecasting are contrasted with the ANN forecast. Using MAE, MSE, and RMSE, one can obtain the predicting capability of these two methods. The findings demonstrate that ANNs are capable of accurately predicting stock market prices when they are trained with enough data, appropriate inputs, and a suitable design. As a result, ANN offers a superior substitute method for daily stock market price forecast.