This project covers predicting the next day's direction of movement for the index of NYSE based on various sets of initial variables. I found and analyzed the relevant data:
Article: CNNpred: CNN-based stock market prediction using a diverse set of variables, Expert Systems with Applications,Volume 129,2019,Pages 273-285, Link: https://www.sciencedirect.com/science/article/abs/pii/S0957417419301915?casa_token= X4geTnwiPW0AAAAA:glja7HHWTT2byFbUuJtn_Sii2oOKfZtQHTX8wOtXQQbLK7ZxAa2 L60LPC0EFBGwlJO8bliq9
Authors: Ehsan Hoseinzade, Saman Haratizadeh For this prediction task, I had 81 potential predictors of the Closing Price for representing each day of each index. Some of these variables are index-specific while the rest are general economic variables and are replicated for every index in the data set. This set of predictors can be categorized in eight different groups: primitive variables, technical indicators, world stock market indices, the exchange rate of U.S. dollar to the other currencies, commodities, data from big companies of the U.S. markets, future contracts and other useful variables. This data is from the period of Jan 2010 to Nov 2017. It had 1985 observations, including those with missing values. After omitting them, I was left with 1114 observations. The first 80% of the data is used for training the model and the last 20% is the test data. The value being analyzed was “Close” that is the closing price of NYSE. I dropped the columns “date” and column 59 as it was not identified as the potential predictors