Stock Price Forecasting Exogenous Variables

Provides public stock market information, including prices augmented with exogenous variables, offering a robust resource for advanced financial modelling and data analysis. Stock markets involve the aggregation of buyers and sellers, which represent ownership claims on businesses. Accurate prediction of stock market returns is a very challenging task due to the volatile and non-linear nature of the financial equity markets. This data is crucial for developing highly efficient programmed prediction methods leveraging artificial intelligence and increased computational capabilities.

The data structure for submission purposes includes key identifiers and the required prediction variable:

The underlying data file is usually supplied in CSV format. The sample submission file is 54.33 kB in size and contains 2 primary columns. The structure is suitable for submitting forecasts, detailing 4,223 valid records.

This resource is ideally suited for applying data science skills to financial challenges. Primary use cases include developing machine learning models for forecasting future closing prices for the 104 listed stocks over the specified two-month horizon. It serves as an excellent benchmark for testing algorithms designed to handle the non-linear volatility present in equity markets.

The data covers variables related to a public stock market, specifically focusing on 104 individual stocks. The prediction range specified covers the period from 1st November 2019 to 31st December 2019.

CC0: Public Domain

Intended users include data science practitioners and analysts looking to improve forecasting accuracy in financial contexts. Stock market participants and investors can utilise the derived insights to inform their decision-making processes, as successful trading relies on foresight into stock performance.

Original Data Source: Stock Price Forecasting Exogenous Variables