https://francis-press.com/uploads/papers/zRXxfyCTUTCMuPGXWJ61aMmIjjXtcwXB7vzsgEFc.pdf

Utilizing historical data from the Quandl API, this experiment investigates use of Convolutional Neural Networks (CNNs) in stock price prediction. The information for Microsoft Corporation (MSFT) spans dates January 1, 2013, and May 18, 2018. high, low, open, close prices are scaled, input sequences length 6 created order to capture temporal dependencies. Dense layers, 1D convolutional max pooling, dropout regularization all components CNN architecture. Mean squared error (MSE) loss Adam optimizer used train model. absolute (MAE) root mean square (RMSE) assess performance. To evaluate model's convergence generalization, losses errors training validation examined. For visual purposes, anticipated actual contrasted. results shed light on how well CNNs anticipate prices, assisting investors financial institutions making wise choices.