In the last several posts I have been discussing the different frameworks that can be used to make predictions on bitcoin. One model that I recently discussed was to use the recurrent neural network (RNN), and that post can be seen here:- https://medium.com/@tracyrenee61/can-you-use-a-keras-rnn-to-predict-bitcoin-prices-9f5771017913
I have recently read that the long short term memory (LSTM) is a modification of the RNN, so I wanted to see for myself how these two models differ from each other.
The LSTM is an improved version of the RNN designed by Hochreiter and Schmidheuler. LSTM is well suited for sequence prediction tasks and excels in capturing long term dependencies. Its applications extend to tasks involving time series and sequences, LSTMs strength lies in its ability to grasp the order dependence crucial for solving intricate problems, such as machine translation and speech recognition.
Since the LSTM is an improved version of the RNN, I wanted to know if the LSTM outperforms the RNN when making predictions on bitcoin.
