Predicting stock market values is a cool RNN application and there are a number of handy packages available, for instance:

• Drnns-prediction implements deep RNNs using the Keras neural networks library on the Daily News for Stock Market Prediction dataset from Kaggle. The dataset task is to predict future movement of the DJIA using the current and previous day's news headlines as features. The open source code is available at https://github.com/jvpoulos/drnns-prediction.
• Michael Luk wrote an interesting blog posting on how to predict Coca Cola stock volume based on RNNs: https://sflscientific.com/data-science-blog/2017/2/10/predicting-stock-volume-with-lstm.
• Jakob Aungiers wrote another interesting blog posting on LSTM Neural Network for Time Series Prediction: http://www.jakob-aungiers.com/articles/a/LSTM-Neural-Network-for-Time-Series-Prediction.