---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: answer
    dtype: string
  - name: period
    dtype: string
  - name: label
    dtype: string
  - name: symbol
    dtype: string
  splits:
  - name: train
    num_bytes: 3127735
    num_examples: 480
  - name: test
    num_bytes: 797367
    num_examples: 120
  download_size: 1523163
  dataset_size: 3925102
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---
# Dow30 Stock Prediction Dataset

## Overview

Welcome to the Dow30 Stock Prediction dataset! This dataset is designed to assist in predicting stock returns for companies in the Dow Jones Industrial Average (Dow30). It includes essential information about each company, such as news from the last two weeks, basic financial data, and stock prices over the same period.

## Dataset Structure

The dataset consists of the following columns:

1. **prompt:** Information about the company, including news from the last two weeks, basic financial data, and stock prices for the same period. The system prompt is generated using the code provided in the [FinGPT_Forecaster](https://github.com/AI4Finance-Foundation/FinGPT/blob/master/fingpt/FinGPT_Forecaster/prepare_data.ipynb) repository.

2. **answer:** Stock return predictions generated by ChatGPT.

3. **period:** Time period of the data, recorded on a weekly basis.

4. **label:** Indicates whether the stock is predicted to go up or down, along with the percentage change.

5. **symbol:** Stock symbol representing the company in the Dow Jones Industrial Average.