Alternative Data Sets Guide Better Quantitative Analysis

Pete Millington

Pete Millington·– November 19th, 2021

Alternative Data

Alternative data sets are quite sought after in the quantitative investment management world. When everybody is using the same traditional metrics and means, competitive advantage lies in being able to outperform someone with a similar background. In the world of investment management, a novel approach is important to achieve this out performance. Dealing in margins, though, you do not need much of an edge to achieve significant gain. After all, a few basis points of a billion dollars is still a very large number. Having access to alternative data can help provide that margin.

Large financial institutions have started using their considerable resources to get expensive data sets. “Alternative data” is going mainstream as fund managers are projected to spend more than $1 billion this year to beat the market averages and stave off the rise of low-cost passive investing. (Source: www.marketwatch.com) This presents itself as an disadvantage to smaller financial outfits that may lack the resources to spend on data.

However, many new sources exist, some of which are free. For example, a tremendous amount of data is available with Google Dataset Search, a free tool for searching millions of publicly available data sets. With a simple keyword search, Google Dataset users can discover data sets hosted in thousands of repositories across the Web.

Scientific Financial’s Quotient platform levels the playing field. By providing a very flexible data integration and data management facility, it easily allows users to construct an extensive, diverse, and valuable data environment. It is like the international electrical outlet converter so well known to seasoned international travelers. Quotient allows you to work seamlessly with many different contexts and perspectives. If you want to stitch various data sets together, a lot of the messiness (like currencies, time frequencies) is handled for you by Quotient.

Analyst at computer using Quotient for Quantitative Investment Analysis and Alternative Data

Sourcing Alternative Data

Google Dataset notes on their website, that they seek to foster a data-sharing ecosystem that will encourage best practices for data storage and publication. The hope is, that as more dataset repositories use schema.org and similar standards, the variety, and coverage, of datasets will continue to grow.

We tested Google datasets by doing a search and then pulling up different datasets and results, including “Rail cargo traffic in Nigeria 2011-2018.” This data set could provide potential insights into commodity trends during this time.

We then input “tanker” and pulled up “Number of oil tanker arrivals in Singapore 2008-2020.” This data set could provide potential insights into disruptions in oil shipments to Southeast Asia. Have a look to see the possibilities for yourself.

Locating interesting datasets is only the first step; working with large amounts of data is a messy and time-consuming task. Large investment shops can wield a competitive advantage, as they can uniquely afford the expensive resources and infrastructures to tackle the data management problems.

Taking the Plunge

The conventional approach to the market makes it difficult to get close enough to the dark matter to make a difference. By using alternative data, traders and investors can get a better understanding of the market and make more substantial differences in returns. While alternative data can help, it can also lead to problems because the standard data set is inaccurate.

Quotient's capability yields a treasure trove of analytical possibilities for quants to excel and stand out against their peers. Quantitative analysts have more tools at their disposal with which to manipulate data. This arsenal delivers effective experimentation and self-education to produce differentiated, unique, and valuable results. And when you are ready to go live with your trading strategies, Quotient is there to seamlessly manage your production environment for you.

This diversity of technologies and content delivers unique insights, valuable, and actionable opportunities. There is great value hidden in new, alternative information sets. Effective data exploration is all about finding the right mixture of content, tools, techniques and imagination to make new discoveries. Let the data expedition begin!