What is Quantamental Investing®?
This article is the 1st in a series of 3 explaining Quantamental Investing®. Here is a quick map — [I. What is Quantamental Investing®?] >> [II. Quantamental — A Convergence of Intelligence] >>[III. A Quantamental Road-Map to Constructing Portfolios].
Quantamental is a buzzword in today’s world of investing. Compared to fundamental investing (about 100 years old) and quantitative investing (about 50 years old) quantamental is still new. It is a philosophy gaining traction with both mainstream and boutique investment managers, yet it is little understood.
By a simple definition, Quantamental is an agreement between fundamental and quantitative styles of investing. Beyond this, the definitions of the term are colored by each practitioner’s use of the concept. No matter the definition, Quantamental has some definite parts.
At Modulor Capital we practice Quantamental Investing® to create portfolios at both Public Equity and Venture Capital Stage. This article is intended to broaden your view on Quantamental Investing®.
Deconstructing Quantamental
Let us understand Quantamental via- negativa (just to clear vague definitions floating around):
- It is not just the use of ML and AI in investing.
- It is not only the use of Alternate Data.
- It is not just the use of NLP to understand transcripts of an investor-call
- It is not an isolated sentiment analysis.
- It is not just pure data science.
It is a mixture of everything. Quantamental was not a word till a few years ago. It is made up of 2 words:
QUANTitative + FundAMENTAL = QUANT+AMENTAL
So we can surely say that Quantamental uses elements from multiple schools on investment. The three primary schools are:
- Fundamental
- Quantitative
- Technical
(We will skip discussing Macro Investing because it does not usually pertain to stock markets or venture capital and its players are large institutional investors). Let‘s get the background on the three schools.
Fundamental Investing — The School of Valuation
The oldest and the most popular school that has been in practice since before the invention of computers is fundamental. The likes of Benjamin Graham, Warren Buffet, John Templeton, Peter Lynch, and Philip Fisher are the pioneers in this domain.
The human capabilities and financial data of companies are used to arrive at the fair value of a business in comparison to its actual value. Analysts use financial statements, management guidance, etc. to estimate the future performance of a business. In a simple logic:
Fair Price > Actual Price → Buy the stock
Fair Price < Actual Price → Sell or Short the stock
Whether top-down or bottom-up, analysts approach one company at a time and delve deep into a business to estimate its worthiness (like how cheap it is or how much is it expected to grow). Analysts are typically sector specialists and also consider macroeconomic and sector-specific trends to get perspective.
When standardized data is unavailable, fundamental investors use intuition and logic to creatively put together disjoint facts. This is truly an area where the human mind excels. At their core, fundamental investors are good businessmen themselves and make sense of the company’s management actions. Presently, investing in early-stage companies (venture capital) is done in this way because data and facts are less available.
Fundamental investors may analyze a company every quarter when it releases its earnings. The style is essentially used for investing, and not trading.
Fundamental-Quant
With the ease of computing and the rise of databases, Fundamental-Quant Investors evolved. They use proprietary screens of financial ratios to scan a large number of companies. They may estimate a company’s future using tools like Discounted Cash Flow (DCF) models.
Fundamental-Quant investors create a universe of stocks by elimination. They then carefully investigate a business in-depth to select it. A typical Fundamental Equity fund works like this…
- Select size universe (large, mid, small, etc.) and sectors to invest in based on the themes that look promising in the future.
- Run screen to eliminate unqualified stocks basis financial ratios like Debt to Equity Ratio (D/E), Price to Earnings Ratio (P/E), Return on Equity (ROE), Earnings per Share (EPS), Quick Ratio, Working Capital Ratio, etc.
- Examine individual stocks for their earnings, governance, management quality, analyst estimates, future outlook, etc.
- Make a collection of top stocks from different sectors and give them weightage for the portfolio.
Of course, this is not the exact process, and variations exist between investors and managers. Fundamental investing can have different styles like value, growth, contrarian, etc. However, it is mostly a discretionary process.
Role of machines in Fundamental Investing
Machines have entered fundamental investing through techniques like Natural language Processing (NLP). Specialized algorithms quickly understand the manager’s intention on an investors’ call or the mood of an annual report. These data inputs are then used to put context to the prospects of the company.
Machines also help sense public sentiment around a stock by scraping social media feeds, news sites, etc.
Quantitative Investing — The school of Anomalies
Computing power and availability of historical data as a time series gave rise to quantitative investors or “Quants”. Quants use Market data like price, volume, open interest, trading statistics, etc. Famous quants include Jim Simmons and his hedge fund Renaissance Technologies, the genius professors Merton and Scholes, and their infamous partnership named Long term Capital Management (LTCM).
Quants use statistical tools to find and exploit anomalies in the markets which are in the form of Factors (momentum, size, volatility, quality, etc) or Statistical Arbitrages (correlations, mean-reversions, etc.). Models showing promising results over historical data are converted into algorithms for trading and investing. These methods work well on multiple asset classes for trading in microseconds or investing over months and years
Quants do not care for the real-world explanation of their factors or anomalies. If you ask a Quant why the price of the stock fell, she will tell you it has mean-reverted, uninterested whether it was a bad quarter or a business accident. Quantitative strategies can have risk-reward characteristics different enough to be considered an asset class in themselves.
Multi-Factor Models
Investing for quants has moved towards Factors that explain the outperformance of a stock concerning others. Factors are cross-sectional (comparing different stocks over the same parameter). Popular Factors are classified as:
- Fundamental
- Statistical
- Macroeconomic
Further research shows that the first two do most of the explanation of Alpha. A typical Quantitative Investment process will look like this:
- Select a broad universe.
- Create multiple rank lists of stocks using different factors like Size, Momentum, Volatility, Quality, Value, etc.
- Combine these rank lists either sequentially or give weightage to each factor and combine them in an equation and arrive at a final rank list.
- Choose the first N stocks on the list to invest in and give them weights either by rank or by a particular factor (like size) or just equal weight.
Again, this is not the exact and only process used by all investors/ managers. Quant funds are typically factor-based. The ones which trade statistical arbitrage are called — arbitrage funds
Systematic Trading
Quants can use factors to create a set of investing rules or systems. These rules can be used to train neural networks over large sets of data to form complex entry, exit, and money management rules. Neural nets can give different weights to different factors and in effect work like a discretionary human investor.
There is an eerie similarity between a discretionary fundamental investor’s methodology and a systematic investor’s methods. For example, a fundamental investor may want to buy cheap companies. This is the equivalent of the value factor for a quant. The same goes with good accounting practices matching management quality as a factor and so on.
Big data in Quantitative Investing
Quants love data and we are living through a data revolution. Extremely large sets of data are analyzed using machines to reveal patterns invisible to the human eye which are used for creating investing opportunities.
Experts in these areas are often not from a finance background (more likely physics, mathematics, etc). However, finding the right direction to look in does require human intervention.
Fundamental vs. Quantitative — An Age-Old Debate
Analysts determine “absolute valuation”, while Quants looks at “relative valuation”. The first focuses on the depth of a stock, while the second focuses on creating a cross-sectional view amongst stocks. While Fundamental Investing is mostly Discretionary (where the fund manager takes a call on which stock to invest in), Quantitative investing is Systematic (algorithms decide, people follow).
However, neither Fundamental Investors predict when the value of a stock will unlock, nor can Quants say when the anomaly will be realized (unless it is an expiry-based arbitrage).
Timing is the domain of the third school of investing.
Technical Investing — The school of Patterns
Technical Investing relies on time-based price patterns considered over sliced timeframes (like 5-minutes, 15-minutes, 1-hour, 1 day, 1-week, and so on). Chartists plot price against time and predict price action by studying patterns visible to the trained eye. The father of technical analysis is Charles Dow (after whose name the DJIA Index is named).
For a long while, fundamental investors and academia disregarded the technical school as a pseudoscience. Yet, practitioners of technical analysis swear by it and have survived to make money.
With the availability of sophisticated computing, analyzing time series of price transforms became the basis of Technical investing. Indicators such as Moving Average, MCAD, Stochastic, etc. can quickly be programmed and plotted along with the price for traders and investors to make buy or sell decisions. If many indicators are combined along with rules on what to do when an indicator goes off, technical analysis can too be systematized.
Technical Analysis is scalable from a few seconds to investing timeframes of years. Statistics can predict the reliability patterns. The technical school also has both discretionary and systematic investors.
Fractal Geometry
Benoit B. Mandelbrot, considered the father of Chaos theory, describes market movements in terms of randomness and fractals. He found that by scaling in or out, over various time frames, a price chart shows the same features across asset classes (the features of cotton futures are the same as gold or, natural gas or, a stock index are the same in an intraday, daily, weekly, monthly chart, etc.). He talked about turbulence, riskiness, timing, price jumps, time compression, uncertainty and bubbles, volatility, and the limited use of “Value”.
For a long time, these ideas were considered heresies in finance. The fundamental and quantitative schools did not quite agree with these concepts (much like Technical Analysis itself).
While the intellectual debate is still on, complex mathematical and geometrical studies have evolved (with difficult names) which are used in identifying price patterns and timing markets. Technical analysis with all the trading software is the easy domain of every new trader. On the other hand, fractal geometry is exclusive to the math and physics Ph.D. sorts. It is little understood and less popular.
Quantamental — A Multidisciplinary Approach
The investment management industry is not full of purists. Successful investment managers are often cross-disciplinary practitioners. For example, it is common for a value fund manager to use a 200 Day Moving Average for time entries.
Quantamental Investing® is multidisciplinary. It intends to get the best results at low costs using whatever tools available.
Quantamental investors cross over different capabilities, data, and tools to make better decisions. It is a space where humans become comfortable understanding a data-based machine’s decision by finding a reason for it. This gives them the confidence to act upon the analysis (eventually the goal is to make money and that is done by human decision, whether it is to take a position or to deploy an automated trading machine).
Let us understand this with an example when the market falls and the price of a stock falls with it. How does each school look at it:
- Fundamental investors find cheap valuations and a low P/E ratio. If the reasons for the business to continue to hold, they buy at or below the intrinsic value with a margin of safety.
- Quants find a sharp divergence between current and historical prices, which is likely to mean-revert since price typically spends very little time at extreme levels. The value factor will make the stock attractive.
- A Chartist observes a U or a V-shaped bottom formation and a reversal pattern and calls it a buy.
- Alternate data shows that business is still in full swing because trucks are flowing in and out at the same pace.
Combining the reasons to buy the stock gives the quantamental investor higher confidence to take a position. By simply overlapping multiple analyses the accuracy increases many-fold, giving the investor risk adjusted returns.
Modulor Advisory Services is a Securities and Exchange Board of India (SEBI) Registered Investment Adviser (RIA) with license number INA100015115.
