Where Models Meet the Unknown: Lessons from Rosenberg and Quantamental Investing
Mona Zhang, CFA
March 25, 2025
As a fundamental investor, I’ve always believed that navigating market cycles and creating long-term returns comes from a deep understanding of business fundamentals. Over time, however, I’ve come to recognize that quantitative tools play a unique and valuable role in portfolio management—particularly in risk assessment and interpreting market behavior.
Sometimes, we don’t just lose a scholar—we lose a way of seeing the world. When I learned of Barr Rosenberg’s passing, I felt a deep sense of reflection. Having used Axioma Inc. Model—known as a key competitor to the Barra Model created by Rosenberg—I decided to write this short piece, both in memory of him and to revisit my own exploration of Quantamental investing and risk factor models.
💡 Who Was Barr Rosenberg?
Rosenberg is regarded as one of the pioneers of modern quantitative investing and risk factor modeling. The Barra Model he created is still widely used by institutional investors and asset managers worldwide to help identify and quantify portfolio risk factors, adding structure and discipline to risk management.
What is less known is that Rosenberg was also a devoted Buddhist, someone who consistently approached the world’s uncertainty—and the market’s unpredictability—with humility and reverence. As a fellow practitioner, this philosophy resonates deeply with me.
How Quantitative Tools Complement Fundamental Investing
In my own investment practice, fundamental research has always been the foundation—understanding a company’s value proposition, business model, competitive advantages, growth potential, culture & people, and financials (a framework that Trunity Partners Ltd. applies in our everyday analysis).
Over the years, I have also explored how to integrate quantitative tools, especially risk models, into portfolio management and position sizing decisions.
At the time, I worked primarily with Axioma Model, a direct competitor to Rosenberg’s Barra Model. While I lean toward fundamental analysis, I must admit that quantitative models opened up my eyes to another dimension: the hidden world of portfolio-level “factor risks.”
1. Identifying Hidden Risk Exposures
These tools are particularly helpful for managing diversified portfolios with larger AUM, especially those benchmarked against indices. They enable us to systematically uncover factor exposures—Value, Growth, Momentum, etc.—and even assess how sensitive the portfolio might be to inflation or interest rate changes. And more so, to help us quantify the sensitivity to various economic factors, instead of relying on PM’s “gut feeling”.
💡 What Are Style Factors?
Style factors are widely used risk factors in quantitative investing. They represent systematic return drivers such as Value, Growth, Momentum, Low Volatility, and Quality. Quant investors use models to measure a portfolio’s exposures to these factors and assess whether the portfolio is “balanced” in terms of risk style and potential factor biases.
To borrow a food analogy:
Fundamental investors are like chefs—they focus on the ingredients of the meal: how much vegetables, protein, grains, or dairy (representing industries, companies, regions, or business models) make up the portfolio.
Quantitative investors, however, are like nutritionists—they focus on the nutritional content of the meal: how much protein, carbohydrates, fats, or micronutrients it contains (corresponding to style factors like Value, Growth, or Momentum).
Both perspectives aim to assess whether the “meal”—or portfolio—is well-balanced. They simply approach it from different angles: one looks at ingredients, the other looks at nutrients. Both can be complementary in building a healthy investment strategy.
2. Optimizing Position Sizing Through Factor Tilts
More importantly, models provide guidance on adjusting position sizes based on the portfolio’s risk tolerance—suggesting where to add or trim positions to reduce volatility or factor concentration.
A Case That Left a Lasting Impression: How I See Models Today
In 2021, while managing an internal portfolio, I found myself navigating a market dominated by highly valued U.S. tech stocks. The portfolio carried significant Growth and Momentum exposure at the time.
The risk model I was using flagged this imbalance and recommended increasing exposure to Broadcom , pushing the position to the upper limit of the portfolio’s risk tolerance.
Back then, Broadcom wasn’t yet part of the AI hype cycle—it was still largely a semiconductor company focused on network switches and software acquisitions. But its strong Value tilt (or to say, reasonable valuation) provided a natural hedge against the portfolio’s Growth and Momentum risks.
In addition, starting mid-2021, a lot of growth stocks’ valuations had far exceeded their historical average and a potential steep rate hike by the fed was the most discussed risk factor among institutional investors. All of the above would make a fundamental investor worry about the markets.
I followed the risk model’s recommendation and increased the weight of Broadcom to where the risk model suggested; the stock went on to significantly outperform the markets during the 2022 market downturn.
Like many active fundamental investors, I used to rely on experience or gut feel when sizing positions—whether a top holding should be 7% or 6.5%. But in this case, the model helped me make that decision more quantitative, systematic, and risk-aware.
This experience reaffirmed my view: models don’t replace investment judgment—but they are a critical second layer of defense.
That said, the other side of a coin might be true too.
🧩 When Models Break: Historical Lessons
History would also reminds us that models, while useful, don’t always perform under all market conditions. A clear example occurred between 2018 and 2020, when the Value factor consistently underperformed the Growth factor.
During this period, declining yields and massive liquidity drove investors into high-growth technology companies. These firms saw resilient earnings and cash flows—especially fueled by remote work trends during COVID—pushing Growth stocks to outperform without the usual mean reversion dynamic.
Another iconic example is the collapse of LTCM in 1998. Founded by Nobel laureates and backed by sophisticated models, LTCM collapsed when over-leveraging and blind reliance on models met the harsh reality of market dislocation and tail risks.
So for quant investors, a bit of a fundamental perspective could be helpful in understanding the limitation of models.
Risk Can Be Measured, But Returns Must Be Predicted
From factor underperformance during 2018-2020 to the collapse of LTCM, these episodes all serve as reminders: “Models can decompose risk—but they cannot predict returns.”
From my experience, the hardest part of any model is not calculating risk but estimating a company’s long-term return.
Models can accurately compute volatility, covariance, beta, and factor exposures—but they can never replace an investor’s qualitative judgment about a company’s growth potential, competitive position, or leadership execution. Especially when it comes to culture, values, and other “intangibles,” no model can quantify these.
In fact, it’s these qualitative insights that often determine whether a company can adapt and thrive in a fast-changing world.
For me, understanding the “people and culture” behind a company is the most difficult—and most valuable—part of investing.
The Buddhist Side of Rosenberg
There’s another side to Rosenberg that I find especially meaningful—he also taught at a Buddhist institute in Berkeley for nearly 40 years.
Though I never had the chance to attend his classes, I understand why he approached uncertainty with such humility. In Buddhism, impermanence is one of the most fundamental truths about the world.
Rosenberg’s investment philosophy never portrayed blind faith in models; instead, it reflected a clear-eyed understanding of their limitations. Models are tools, but long-term investing requires human insight and the ability to cut through short-term noise to see the essence of a business.
This has been a belief I continue to practice in my own career.
🖋️ Author’s Note
The world of investing is full of uncertainty and surprises. But lasting success depends not just on models, but on deep insight into businesses, markets, and people.
Rosenberg’s legacy taught me more than quantitative discipline—it reminded me to stay thoughtful and humble when the world defies easy answers.
👉 So a quiet reminder to myself, and to those on the long-term investing path:
Stay hungry, stay humble—and stay curious.
Disclaimer: This article is for informational and educational purposes only and does not constitute financial, investment, legal, or tax advice. The views expressed are my own and do not necessarily reflect those of Trunity Partners Ltd. or its affiliates. Any references to specific assets, historical events, or individuals are for illustrative purposes and do not imply endorsement or prediction of future performance. Readers should conduct their own due diligence or consult a licensed advisor before making investment decisions.
