In this chapter, you'll be introduced to how to predict a timeseries composed of real values. Specifically, we will predict the stock price of a large company listed on the NYSE stock exchange, given its historical performance.
In this chapter we will look at:
- How to collect the historical stock price information
- How to format the dataset for a timeseries prediction task
- How to use regression to predict the future prices of a stock
- Long short-term memory (LSTM) 101
- How LSTM will boost the predictive performance
- How to visualize the performance on the Tensorboard
Each of these bullet points is a section in this chapter. Moreover, to make the chapter visually and intuitively easier to understand, we will first apply each technique on a simpler signal: a cosine. A cosine is more deterministic than a stock price and will help with the understanding...
