Skip to main content
About HEC About HEC
Summer School Summer School
Faculty & Research Faculty & Research
Master’s programs Master’s programs
Bachelor Programs Bachelor Programs
MBA Programs MBA Programs
PhD Program PhD Program
Executive Education Executive Education
HEC Online HEC Online
About HEC
Overview Overview
Who
We Are
Who
We Are
Egalité des chances Egalité des chances
HEC Talents HEC Talents
International International
Campus
Life
Campus
Life
Sustainability Sustainability
Diversity
& Inclusion
Diversity
& Inclusion
Stories Stories
The HEC
Foundation
The HEC
Foundation
Summer School
Youth Programs Youth Programs
Summer programs Summer programs
Online Programs Online Programs
Faculty & Research
Overview Overview
Faculty Directory Faculty Directory
Departments Departments
Centers Centers
Chairs Chairs
Grants Grants
Knowledge@HEC Knowledge@HEC
Master’s programs
Master in
Management
Master in
Management
Master's
Programs
Master's
Programs
Double Degree
Programs
Double Degree
Programs
Bachelor
Programs
Bachelor
Programs
Summer
Programs
Summer
Programs
Exchange
students
Exchange
students
Student
Life
Student
Life
Our
Difference
Our
Difference
Bachelor Programs
Overview Overview
Course content Course content
Admissions Admissions
Fees and Financing Fees and Financing
MBA Programs
MBA MBA
Executive MBA Executive MBA
TRIUM EMBA TRIUM EMBA
PhD Program
Overview Overview
HEC Difference HEC Difference
Program details Program details
Research areas Research areas
HEC Community HEC Community
Placement Placement
Job Market Job Market
Admissions Admissions
Financing Financing
Executive Education
Home Home
About us About us
Management topics Management topics
Open Programs Open Programs
Custom Programs Custom Programs
Events/News Events/News
Contacts Contacts
HEC Online
Overview Overview
Degree Program Degree Program
Executive certificates Executive certificates
MOOCs MOOCs
Summer Programs Summer Programs
Youth programs Youth programs
Article

Big Data and Investment Returns: Insights by HEC Professor Hugues Langlois

Finance
Published on:

Hugues Langlois, Assistant Professor of Finance at HEC Paris, shares the practical applications of his research about big data methodologies in finance, and how it provides a new tool to measure expected returns in international stock markets. A big step in the world of finance for asset managers and risk managers.

 

What are the applications of big data in finance?

In recent years, we have been hearing a lot about applications of machine learning, deep learning, and artificial intelligence in fields such as robotics and medicine. But there are also many applications in finance. 

For example, we use in a recent research project a big data methodology to extract the investment return investors should expect when investing in stocks all around the world. 

An asset-pricing model is an economic model that explains why some assets, such as stocks, have higher returns on average than others.

Many finance researchers strive to find the best asset-pricing model. An asset-pricing model is an economic model that explains why some assets, such as stocks*, have higher returns on average than others. The hope is that by explaining why average returns differed in the past, we may measure the return we expect to realize on investments in the future (i.e., their expected return). 

Why is a good asset-pricing model useful for investors?

Investors equipped with a state-of-the-art model can identify good investment opportunities in the market and form better-performing portfolios. Consequently, finance academics and fund managers are engaged in an arms race to find better models. Indeed, sophisticated asset managers like AQR, DFA, Man Financial, Arrowstreet or Amundi Asset Management either employ or work closely with top academic researchers. 

A good asset-pricing model also has tremendous importance in corporate finance applications. Indeed, the returns investors expect when buying a stock is also the cost of equity capital a corporate manager has to take into consideration when making a decision about investing in a new project.

Our methodology: using stock portfolios, machine learning, and statistics

In academic finance, we have traditionally tested our different theories using portfolios in which stocks with similar characteristics are grouped together. For example, we use portfolios to address long-standing questions in finance: do riskier stocks have higher returns on average than safer stocks or do stocks of small companies have higher average returns than stocks of large companies?

Up to now, using stock portfolios to test our theories avoided two key hurdles faced when using individual stocks. The first set of hurdles are statistical biases, meaning that you may not necessarily obtain the expected returns you want to estimate even if you had a large amount of data. A second set of hurdles concerns computational difficulties. There are measures that in the past were simply too hard to compute when dealing with thousands and thousands of stocks.

For example, any student in a Statistics 101 course can compute a measure of how two variables co-move together. But the same computations on tens of thousands of stocks is challenging, if not impossible.

Big Data and investment returns @James Thew
Big data methodologies help to measure expected returns in international stock markets. Photo @James Thew/AdobeStock

In recent research with my colleagues Olivier Scaillet and Ines Chaieb at the University of Geneva, we use recent developments in machine learning and statistics to overcome these issues. We use an exhaustive list of more than 60,000 individual stocks in 46 countries over a more than three-decade period to extract information on key issues in stock markets: what are the main factors that drive expected stock returns and how do these expected returns change over time?

Can you tell us more about the practical applications of this research?

A recent survey** done by the index provider FTSE Russell found that almost four out of five asset managers plan to adopt, or have adopted a smart beta strategy, meaning that they follow strategies that are systematic, rule-based, transparent, and well-documented.

Our big data methodology is directly relevant for their needs: We can identify which, and how many of these smart beta strategies they need in their portfolio, and we can measure their current expected returns. 

A good asset-pricing model is the cornerstone of asset management, risk management, performance evaluation, and corporate finance. Our research provides a new tool to measure expected returns in international stock markets.

Understanding the expected return on a stock…

Understanding and measuring the determinants of expected returns—also called risk factors—in stock markets is crucial to allocate investment portfolios, evaluate the performance of fund managers, and obtain the cost of capital for firms. The expected return on a stock can be understood as coming from two components: (1) its exposure to risk factors, and (2) the expected returns on these risk factors.

…for asset managers 

Let’s start with an application to asset management. The viability of a smart beta or a multi-factor strategy crucially depends on our ability to identify and combine risk factors in a portfolio. A better understanding of the mechanisms at work brings benefits in terms of both return and risk.

In our research, we use information extracted from a large database of individual stock returns, firm characteristics, and macroeconomic variables to address three specific needs faced by asset managers. 

First, we determine whether a candidate set of risk factors (or smart beta) fully captures the structure of co-movements between stock returns. For example, should a fund manager add a profitability strategy if he or she already follows a value and momentum strategy?

Second, we measure the time-varying compensation for each risk factor. For instance, we can determine whether the compensation for holding stocks of small companies is the same as it was 20 years ago. We show how the compensation investors receive when investing in the broad market, size, value, momentum, profitability, and investment factors, either constructed at the country, regional, or global level, changes over time. 

Finally, we test for the presence of pockets of arbitrage opportunities in stock markets. While this point is more technical, it is nonetheless important for asset managers. This issue addresses questions such as: If I control for exposures to well-known risk factors, can I find a strategy that delivers positive return with very little risk?

…for risk managers 

Our research is also important for risk managers and allocators of capital. For example, a good asset-pricing model serves as a benchmark to measure active managers’ risk and outperformance. Therefore, identifying risk factors in international stock markets is important for performance evaluation and risk management of stock portfolios. 

Finally, our research is important for corporate managers as the cost of equity financing is a crucial input when evaluating the value of investing in a new project, of acquiring a company, etc. 

A good asset-pricing model is the cornerstone of asset management, risk management, performance evaluation, and corporate finance. Our research provides a new tool to measure expected returns in international stock markets.

 

*We examine stocks in this text, but asset pricing models are used for any asset classes.

**See "Smart beta: 2018 global survey findings from asset owners"

Interview based on the academic paper “Time-Varying Risk Premia in Large International Equity Markets”, by Hugues Langlois (HEC Paris), Ines Chaieb and Olivier Scaillet, 2018 (working paper).

Related content on Finance