liveBook · Manning

Market data

As explained in Financial AI in Practice

Data describing financial markets, including prices, trading volumes, and market activity.

A quantitative trading strategy may use market data—prices and trading volumes—to look for predictive signals. Before modeling, the data must reflect what actually happened in the market, including corporate actions and securities that later disappeared. section 10.2.1

Three details can change the meaning of a signal:

  • Adjust prices for stock splits, dividends, and rights offerings. Otherwise, a split can look like a sudden crash. section 10.2.1
  • Include bankrupt and delisted companies in historical tests. Excluding them creates survivorship bias and can make a strategy look more profitable than it is. section 10.2.1
  • Choose a representation that preserves the needed information. Candlesticks show open, close, high, and low prices; closing-price line graphs simplify trends but omit intra-period detail. section 10.2.1

How Financial AI in Practice covers it

How deep it goes
Explains data-quality pitfalls and chart representations, but stops short of a data-processing implementation.
Examples
Uses a two-for-one stock split mistaken for a crash and contrasts candlesticks with closing-price line graphs.
What you can do after
Evaluate/understand it by checking price adjustments, delisted-security coverage, and information lost through chart representation.
How it connects
Connects market data to corporate actions, survivorship bias, quantitative strategies, and chart-based data.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Financial AI in Practice. The section links point to the parts of the book it draws on.