Information-Driven Bars for Finance

Information-Driven Bars for Finance

An Informal Review of Tick Imbalance Bars. As you may know, raw financial data is made of ticks, which represent market transactions.


As you may know, raw financial data is made of ticks, which represent market transactions. Every market transaction is characterized, among other things, by a date, a time, a price and the number of contracts or shares that have been exchanged.

Image for post

An example of raw tick data taken from a securities database

To represent financial time series you normally use bars, which are aggregations of ticks. But there are several ways to aggregate information; for this reason, many kinds of bars exist (time bars are the most common).

Observe the market for 5 minutes, note down the first observed price (open), the lowest price (low), the highest price (high) and the price of the last tick that belongs to the chosen time interval (close) and you will have built a 5 minute bar!

However, markets, unlike humans, do not follow a time clock. They do not process information at constant time intervals. They rather do so event by event, transaction by transaction. So, what if we abandoned the time clock and started sampling market prices using a different logic? We could use the volume clock to build volume bars for example; or we could build tick bars and sample prices every time we observe a given amount of new transactions. Volume and tick bars are known sampling methods and I do not think they deserve an entire article as you can find plenty of information on the web.

But, what happens if we use an information clock? What if we sampled prices every time an unexpected amount of information comes to the market? Well, this is precisely the idea behind information-driven bars. Interesting, isn’t it? We are going to introduce the simplest kind of information-driven bars: tick imbalance bars. Marcos Lopez De Prado presents them in his book Advances in Financial Machine Learning (AFML).

information-driven-bars trading de-prado-advances-in-ml financial-data-sampling data visualization

Bootstrap 5 Complete Course with Examples

Bootstrap 5 Tutorial - Bootstrap 5 Crash Course for Beginners

Nest.JS Tutorial for Beginners

Hello Vue 3: A First Look at Vue 3 and the Composition API

Building a simple Applications with Vue 3

Deno Crash Course: Explore Deno and Create a full REST API with Deno

How to Build a Real-time Chat App with Deno and WebSockets

Convert HTML to Markdown Online

HTML entity encoder decoder Online

Visual Analytics and Advanced Data Visualization

Visual Analytics and Advanced Data Visualization - How CanvasJS help enterprises in creating custom Interactive and Analytical Dashboards for advanced visual analytics for data visualization

Visual Analytics Services for Data-Driven Decision Making

Visual Analytics Services for Data-Driven Decision Making - Visual analytics consists of data analysis methods and interactive visualization algorithms for analytical reasoning for actionable insights

Visualization Best Practices for Data Scientists

Visualization Best Practices for Data Scientists. Disclaimer: The ideas presented in this article are from the book: Story Telling With Data by Cole Nussbaumer Knaflic.

How to Visualize High Frequency Financial Data Using Plotly and R

This article I show how to use the Plotly package to visualize financial data in high frequency using R. To perform analysis and…

Applications Of Data Science On 3D Imagery Data

The agenda of the talk included an introduction to 3D data, its applications and case studies, 3D data alignment and more.