Stock market analysis has long been a fascinating field of study, not only for investors but also for data analysts and machine learning aspirants. 

The regression problem is often illustrated with stock prices in the Machine Learning discipline. Predicting the price of a company's shares by analysing historical data is an example of how machine learning may be put to use. The first stage in constructing such a system is, of course, to collect the necessary stock market data.

It is necessary to have historical stock data to analyse the stock market. With the advent of financial technologies (FinTech) and the drive towards inclusive finance, many free-market data sources are now available online. 

In 2023, you can use the following packages to retrieve historical data for single or several stocks. 

Alpha Vantage 

Since Alpha Vantage was established in 2017 and is a product of the Y Combinator accelerator, it might be considered a new player in the industry. Alpha Vantage is developing application programming interfaces (APIs) to facilitate collecting, organising, and sharing data from various financial information sources. In addition, they employ supervised machine learning algorithms that steadily enhance data quality in response to user input. The more people who use Alpha Vantage, the smarter it gets. It permits more data to be added to the cloud thanks to the ongoing feedback of data. They help with stock market history, foreign exchange data, cryptocurrency data, technical indicators, and industry results.

Yahoo Finance

Yahoo! Finance is an integral part of Yahoo's ecosystem. Stock quotations, press announcements, financial reports, original material, financial news, data, and opinion make it the most visited business news website in the United States. In addition, cryptocurrency, fiat currency, commodity futures, stock and bond market data, fundamental and option data, market analysis, and news are all available from this source. 

Quandl 

In 2011, Tammer Kamel established Quandl to create a new Wikipedia for quantitative information. They don't collaborate with suppliers but rather gather data via scraping websites and other available resources. As a result, Quandl has evolved into a "search engine" for quantitative information. Furthermore, all of the information provided by Quandl is in a standardised format, allowing you to quickly locate the information you need and begin using it immediately.

Pandas DataReaders

Pandas-DataReader is an open-source Python module for working with and analysing data. Therefore, the Pandas-DataReader subpackage aids the user in constructing data frames from numerous online resources. Naver Finance, the Bank of Canada, Google Analytics, Kenneth French's data repository, and 16 more are just some sources users can connect to using this tool. 

Pandas DataRedears is an API in the PyData stack that permits access to many different data sources, but it is not a data source in and of itself. Instead, the data will be downloaded as a Pandas dataframe. 

Getting Data from FRED

FRED, managed by the Research Department of the Federal Reserve Bank of St. Louis, contains over 765,000 economic time series from 96 sources. The DataReader API gives us access to this massive data by specifying the symbol category and the indicator for which we require the data.

Image source: Unsplash