Decision tree learning - predicting the direction of stock movement

Stock trading is one of the most challenging problems statisticians are trying to solve. There are multiple technical indicators, such as trend direction, momentum or lack of momentum in the market, volatility for profit potential, and volume measures to monitor the popularity in the market, to name a few. These indicators can be used to create strategy to high-probability trading opportunities. Days/weeks/months can be spent discovering the relationships between technical indicators. An efficient and less time-consuming tool, such as a decision tree, can be used. The main advantage of a decision tree is that it is a powerful and easily interpretable algorithm, which gives a good head start.

Getting ready

In order to perform decision tree classification, we will be using a dataset collected from the stock markets dataset.

Step 1 - collecting and describing the data

The dataset to be used is the Bank of America's daily...