Ideas with Impact
UNB Faculty of Management

Market Predictability - New research to improve investment decision making

Author: Faculty of Management

Posted on Feb 23, 2022

Category: Faculty

The next generation of investors coming from UNB’s faculty of management may have better decision-making skills thanks to the research of Dr. Azam Shamsi. Understanding how markets respond to economic conditions around the world to predict future stock prices is a question she addresses in her article, “Do Financial variables help predict the conditional distribution of the market portfolio?,” published recently in the Journal of Empirical Finance.

Shamsi explains, “Predictability of market returns has important implications for asset allocation and risk management, creating a balanced portfolio of stocks, bonds, and cash in a way that maximizes the investor’s satisfaction.”

Her paper investigates predictability of the full density of market returns in a flexible and rich framework. “I do this without imposing restrictive assumptions on the behaviour of market returns over time and by letting the data speak for themselves. Statistical and economic performance of the proposed model is superior to a set of competing models commonly used in practice.”

Shamsi believes the results of her study will be of interest to the finance community. Institutions such as mutual funds and hedge funds can employ these tools to construct better diversified portfolios and increase investors’ economic value. Academics and asset managers can exploit the developed model to study the behavior of securities.

The question about whether available information on macro-finance variables improves market returns forecasts is a very important question for both academics and practitioners, but the findings are mixed, Shamsi observes. “Some articles provide empirical evidence which supports the predictability of market returns. On the other hand, several studies question the forecasting power of the macro-finance variables. This encouraged me to reassess the problem and investigate it from a different perspective using a less restrictive framework.”

Shamsi is currently working on extensions of this model to measuring, modelling, and forecasting market risk by macro-finance variables.

“The extended model allows data-driven shrinkage of the impact of each potential predictor. Besides the ‘predictability’ context which has a lot of room for extension and improvement, and the methodology and tools developed in this work will help me be better equipped to conduct my future research and solve finance problems under more realistic assumptions.”

Shamsi joined the faculty of management in 2018. She teaches finance courses in the Master in Quantitative Investment Management, MBA and BBA programs. Her research on financial econometrics and empirical finance has also been published in the Journal of Financial Econometrics and in a book, Handbook of Mixture Analysis.

Many of the questions she addresses, including those in her recent publication, are great examples for her students when applying theory to real world issues. In the past few years, she has coordinated two capstone projects related to this work for her MQIM students. “Their involvement in these challenges helped them make connections between the course material and real-world problems, which in turn helped them develop their own independent research.”

Photo: UNB finance professor Dr. Azam Shamsi has published her research on market predictability in the Journal of Empirical Finance.

Learn more about UNB’s faculty of management, and the Master in Quantitative Investment Management program.

Media Contact: Liz Lemon-Mitchell