Demystifying MomentumDemystifying Momentum

Why this article

The Adaptive Momentum Portfolio in our PMS, Capitalmind Wealth, completed five successful years of live operation on March 5, 2024, making it India’s longest-running quantitative PMS momentum strategy.

Five years is only a fraction of a typical investing lifetime. On the other hand, five years is an eternity in an age of widely available “1-Year Top Performer” lists.

Traditional fundamental investment strategies lend themselves to top-down macro Commentary, detailed sectoral trends, and in-depth company analysis for each portfolio constituent to explain how and why they work.

Quantitative strategies like Adaptive Momentum depend on statistically defined factors. So, they tend to be less relatable.

We wrote this article for the non-finance investor to understand, jargon-free, from our experience managing a quantitative momentum strategy with real capital.

One Rupee invested in the NSE500 on March 5, 2019, would be worth 2.35 nearly five years later, a 135% increase. In Capitalmind Adaptive Momentum, that rupee would be worth 3.53, an increase of 253%. The NSE500, being an index, assumes no costs, while the Adaptive Momentum nav is not a theoretical model portfolio and is net of actual trading costs, STT, and fees.

Over 1,000 investors, including individuals, families, and corporations, are invested in the Capitalmind PMS Adaptive Momentum portfolio. The investors with the highest IRRs (effective rates of return) on their investments have stayed invested through periods of indifferent performance and continue to add to their investments.

That kind of conviction and patience comes from a conceptual and practical understanding of the underlying investment philosophy.

Think of it like owning an electric car. You don’t need to be able to take it apart and rebuild it. But understanding the essential systems and how they work together tells you how reliably you will reach your destination.

The “essential systems” of Momentum Investing

• How quantitative investment strategies differ from traditional fundamental stock-picking approaches
• The principles underlying Momentum Investing, its strengths and weaknesses
• The likelihood that it’ll sustain and the things to keep in mind

This should give any investor, existing or prospective, a foundational understanding of Momentum investing and whether it makes sense in their investment portfolio.

If anything in the following pages sounds “technical” and conveys false sophistication, it is unintended and a failure on our part. Where applicable, you’ll find numbered links to footnotes for further references if you’re so inclined. Feel free to contact us for clarifications at the coordinates mentioned at the end.

What this article is not: A how-to manual of Momentum Investing because there are plenty of those, and momentum investing is really not that complicated¹.

First, a look at the performance of Capitalmind Adaptive Momentum in the five years since its inception.

Adaptive Momentum 5-Year Performance Overview

The charts in this section show the performance of Capitalmind Adaptive Momentum since its inception in March 2019 alongside that of the Nifty 50 and Nifty 500 benchmark indices. The benchmark indices are Total Return, i.e. they include dividends. Adaptive Momentum numbers take fees, transaction costs and STT into account.

₹100 invested five years ago

Cumulative Equity Curve

Calendar Year Returns

One-Year Rolling Returns Compared

Percentage time leading on One-Year Rolling Basis

Five-Year Performance Summary

Adaptive Momentum’s five-year tenure has seen wildly varying market conditions. Claims of being the best Momentum Investing PMS strategy will, at best, be mercurial. What we do know, is that Adaptive Momentum has delivered strong risk-adjusted returns, albeit not without significant periods of trailing the benchmarks on a one-year rolling basis. It has done so while managing a decent amount of capital, and keeping worst drawdowns to significantly better than the benchmarks, which gives us the conviction on its long-term viability as a core portfolio holding for investors.

Headline numbers out of the way, the rest of this short report will examine the underlying differences between Adaptive Momentum and traditional investing styles, their respective strengths and weaknesses, and hard-won real-world lessons.

The difference between Traditional & Quantitative Investing

“If you want to understand the whole, you must first look at the pieces.” - Aristotle

We’ll start with the smallest unit of fundamental investing, i.e., company analysis and valuation, as the jumping-off point to make sense of how Quantitative or Rule-based investing differs from traditional fundamental investing.

Traditional Investing: Assembling a Mosaic

The graphic below is the typical framework used to analyse companies fundamentally.

The graphic broadly covers the significant inputs to determine that present value².

That fair value is driven by its:

• External Environment: Capital Markets and Competition in its space.
• Internal Capabilities: Quality of Management and the assets it owns and can build.
• Moat (or Sustainable Competitive Advantage³): Capabilities that allow the company to thrive and others cannot easily replicate.

The elements of that diagram combine to help us determine an (approximate) fair value for any company.

Investing, in a nutshell, is about finding companies that the market is currently valuing at less than “fair” value. The implicit assumption is that the market will eventually recognise this higher fair value, enabling your returns.

Every fundamental investing style boils down to this.

Great investors build their reputation on being “more right” than the rest in assembling the mosaic.

Quantitative Investing: Distilling the Essence

If fundamental investing is about synthesising multiple inputs, like assembling a jigsaw puzzle, quantitative investing is about distilling the essence.

Most quantitative approaches are attempts at reducing hundreds, even thousands, of inputs that potentially drive returns down to a small but meaningful and repeatable set.

Quantitative Investing starts with an investment thesis, a “what if” set of rules applied consistently to pick portfolios. Buying stocks with higher relative returns, i.e. broadly called Momentum, is only one such set of rules. Other rules could be about buying the highest Return on Capital companies (Quality), lowest Price-Earnings companies (Value), lowest volatility companies or variations and combinations of those.

To count on a set of rules to do as well or better than a painstakingly chosen portfolio, based on deep fundamental analysis, the rules need to:

• have an enduring underlying economic rationale and
• be rigorously validated by historical data⁴

Neither of the above two criteria alone is sufficient to build a quantitative investment strategy.

“In God we trust. All others must bring data.”

- W. Edwards Deming (Pioneering Statistician)

Buying stocks at lows

“Buy low, Sell high”—the most unarguable investing truism.

Let’s consider a quantitative approach that buys stocks beaten down in price.

We buy 20 stocks from the top 100 by market cap, closest to their 52-week lows, and hold them for a month. Let’s say we have done this for over 16 years, from 2007 to 2024. Before reading further, what would such a strategy do compared to buying and holding the Nifty?

Logic suggests this should do well, given we’d be buying stocks relatively cheaper than their historical prices. The chart says we’d be wrong.

Applying a “buy stocks at lows” strategy would have returned just over half the Nifty. Not a great way to invest.

What if we did the opposite? Buy stocks closest to their 52-week highs and hold for a month. Everything else stays the same.

Buying a stock at 52-week highs is better than buying at its 52-week low strategy by quite a margin. However, even that fails to do better than just buying the index.

You’re thinking, maybe buying at lows works if we held those stocks for longer, like a quarter or even a year. We tried all those variations of a “buy low” strategy, and you’d still do better just buying and holding the index.

In markets, what we think should work and what does are often different. Quantitative investing looks at the world as it is, not as it should be.

Bangladesh Butter Production and the S&P

“Torture the data enough, and it will confess to anything.”

– Ronald H. Coarse (Economist)

In 1993, two analysts at First Quadrant published a paper that said that Bangladesh’s butter production was an almost perfect predictor of the S&P 500 index.

In statistical terms, an R-squared⁵ of 0.99, along with a low p-value⁶, can be interpreted as there is a proven relationship⁷ between the variables.

The two analysts successfully highlighted the absurdity of relying only on statistical techniques bereft of the economic rationale for investment decisions.

One of them, David Leinweber, wrote in a Forbes article⁸ in 2012 that twenty years later, he was still getting calls from investors asking for data on Bangladesh butter production.

Correlation is not necessarily causation. In simpler terms, common sense is essential.

Writing about the pitfalls and abuse of backtests⁹ to make grand claims would be an article by itself, so we’ll leave it at; when it comes to quantitative investing, you will never see a bad backtest.

The flowchart below summarises the process we’ve evolved into at Capitalmind when going from a “what if” hypothesis to deploying real money on a quantitative strategy.

The objective of this gated process is not to be seduced by shiny new objects or noise pretending to be signal. Most ideas that sound promising in theory do not make it through initial exploration. Several look promising but don’t survive further validation. Many that survive validation show real-world constraints. A small set makes it all the way to justify deploying capital.

Sound quantitative investing relies on a robust, replicable process validated by, but not tailored to the past.

Why Traditional Investing is like Cooking, Quantitative Investing is like Baking

Another useful analogy to make the distinction clear is Cooking versus Baking.

Cooking is an art involving creative combinations of ingredients. A chef uses intuition and experience to create a unique dish. The dish itself can be slightly different each time they make it.

Baking is a precise science involving following a series of precise tested steps. A tiny variation can drastically affect the final product’s taste and texture. A baker’s creativity is applied to developing and testing those steps to create something unique.

	Traditional Investing	Quantitative Investing
Inputs	Wide: Qualitative (management quality) and quantitative	Deep and Narrow: Extensive amount of historical data
Process	Based on interpretation: Same information can result in contrasting decisions	Objective decisions: Only works if applied consistently
Strengths	Deeper understanding of investments can identify undervalued opportunities that others might miss	Reduces emotional bias, more efficient and scalable, diversification is built-in
Weaknesses	Time-consuming, susceptible to biases, skill and expertise-dependent	Can overlook nuances that don’t fit the model. Dependent on effectiveness of chosen factors
Risk	Managed through discretion	Managed systematically

Momentum Investing is one such type of Quantitative investing.

Understanding Momentum Investing

“A stock that has been moving up strongly for a while is likely to continue doing so a little bit longer. That’s the core idea. The rest is details” – Andreas Clenow (Stocks on the Move)

Momentum Investing, in a nutshell, is a strategy that picks stocks that have had higher relative returns over the recent past and holds them for a defined period. Historical and live performance reviews have shown Momentum investing outperforms buying and holding the market index.

How can such a simplistic strategy offer long-term outperformance versus the market?

Momentum is one of the most well-explored investing phenomena in global finance going back several decades.

Globally Researched by academics and practitioners

The starting point to explore Momentum investing is a 1993 paper¹⁰ by two finance professors at UCLA, Jegadeesh Narasimhan and Sheridan Titman.

They tested various portfolios built using a “J-month / K-month” strategy where they picked stocks based on past J-month returns and held them for K months after portfolio formation. J and K varied from 3 to 12 months in three-month increments. The research tested every combination.

Their finding was that all combinations of past returns and holding periods, barring one, outperformed a basic buy-and-hold strategy net of costs.

“Strategies which buy stocks that have performed well in the past generate significant positive returns over 3- to 12-month holding periods"

In 1999¹¹, Tobias Moskowitz of the Chicago Graduate School of Business and Mark Grinblatt of Anderson School of Business, UCLA, examined the industry component in a stock’s Momentum on 32 years of data from 1963 to 1995.

“Industry momentum investment strategies, which buy stocks from past winning industries and sell stocks from past losing industries, appear highly profitable, even after controlling for size…”

Another paper¹² by Tobias Moskowitz, Yao Hua Ooi, and Lasse Hej Pedersen expanded the scope of examining Momentum to over 25 years of data on five dozen diverse financial instruments, including country equity indices, currencies, commodities, and sovereign bonds.

“We document significant “time series momentum” in equity index, currency, commodity, and bond futures for each of the 58 liquid instruments we consider. We find persistence in returns for 1 to 12 months that partially reverses over longer horizons, consistent with sentiment theories of initial under-reaction and delayed over-reaction. A diversified portfolio of time series momentum strategies across all asset classes delivers substantial abnormal returns with little exposure to standard asset pricing factors and performs best during extreme markets.”

A 2017 paper¹³, “A century of evidence on trend-following investing”, by Brian Hurst, Yao Hua Ooi and Lasse Pedersen, considered monthly returns for 67 markets across four major asset classes: 29 commodities, 11 equity indices, 15 bond markets, and twelve currency pairs.

They constructed equal-weight combinations of 1-month, 3-month, and 12-month time-series momentum strategies from Jan 1880 to Dec 2016, 136 years, rebalanced monthly.

“Even assuming a 2-and-20 fee model (2% of AUM as fixed fee and 20% of gains), a momentum-based strategy outperformed net of costs in every decade from 1880 to 2016, with a minimum outperformance of 2.6% from 1880 to 1889 and a maximum of 15.1% from 1970 to 1979.”

A comprehensive review titled ‘Fact, Fiction and Momentum Investing’¹⁴ by Cliff Asness, Andrea Frazzini, Ronen Israel, and Tobias Moskowitz of AQR Capital made this telling assertion:

“The existence of Momentum is a well-established empirical fact. The return premium is evident in 212 years (yes, this is not a typo, two hundred and twelve years of data, from 1801 to 2012) of US equity data, dating back to the Victorian age in UK Equity data, in more than 20 years of out-of-sample evidence from its original discovery, in 40 other countries, and more than a dozen other asset classes.”

Finally, the most telling validation of Momentum as an investment strategy came from Dr. Eugene Fama, Robert R. McCormick Distinguished Service Professor of Finance at the University of Chicago Booth School of Business. His biography page on the Chicago Booth website refers to him as the “father of modern finance”. He is best known for his work on the Efficient Market Hypothesis (EMH), for which he and his co-authors were awarded the Nobel Prize for Economic Sciences.

EMH, in a nutshell, says capital markets are efficient in processing information. Stock prices at any time are based on correctly evaluating all available information. i.e. In an efficient market, prices fully reflect available information. And if prices reflect all information at any time, there is no way to profit from identifying mispricing.

In 2008, Prof. Fama published a paper titled “Dissecting Anomalies¹⁵”, which explored returns from specific investment strategies that did not conform to the Efficient Market Hypothesis. In this paper, he and his co-author Kenneth R French, based on an analysis of 42 years of data from 1963 to 2005, concluded:

“Anomalous returns associated with net stock issues, accruals, and momentum are pervasive.”

Years later, he referred to Momentum as “the premier anomaly” that persists even as others fell away.

Momentum in Indian markets

In 2019, we did our study into Indian markets¹⁶.

It did not take a stretch to see that Momentum has worked well in India. We examined a few variants of a momentum investing strategy, Naive Momentum, along with volatility-adjusted versions. All momentum portfolios comfortably outperformed a Nifty Buy-and-Hold strategy. Adding volatility-adjusted Sharpe return ratios helped improve the risk-return profile of the strategy by reducing the impact of sharp market drawdowns on portfolio performance. The best momentum investing strategies add refinements to try and include the stocks showing small and consistent increases in price while excluding stocks showing large swings in price.

Momentum has historically worked to deliver excess returns in India.

So, it works. But why?

Researchers have proposed several behavioural explanations for why a strategy that buys past winners and sells past losers outperforms the market consistently over centuries of data across asset classes and countries.

Underreaction (or Anchoring Bias or the Frog-in-the-Pan hypothesis)¹⁷: Investors are inattentive to information arriving continuously in small amounts, so frequent gradual changes attract less attention than infrequent dramatic changes. A company with consistently but gradually improving performance would not attract significant attention. A momentum strategy could enter and ride the gradual price rise.

Disposition Effect¹⁸: A natural inclination to sell winners but hold on to losers to avoid losses. Consistent selling from existing investors acts as a brake on the stock price, increasing the price gradually and, therefore, allowing a momentum strategy to enter.

Cognitive Dissonance¹⁹: News that contradicts investor sentiments causes cognitive dissonance, which tends to get ignored or disbelieved, thus slowing its diffusion into stock prices. Therefore, losers keep getting underpriced while winners keep getting overpriced, sustaining their Momentum.

But then, that’s the theory. And nothing quite replaces lived experience.

Ten Battle-tested Real-World Momentum Investing Lessons

"Experience is one thing you can’t get for nothing."

Oscar Wilde

From under ₹5 Cr in 2019, the Capitalmind Adaptive Momentum portfolio has grown to over ₹750 Cr in assets under management. A few things happened along the way. An indifferent 2019, the pandemic crash and its aftermath in 2020-21, the inflation/rates-led reversion in 2022 and the 2023 market resurgence in the face of major geopolitical conflicts. Some pre-existing beliefs about momentum investing were reinforced, others were challenged leading to modifications. And more ideas fuelled the need for further exploration. What follow are our hard-won lessons from the last five years.

Assumes we know less

Successful fundamental investors make money by knowing something about a company that the people they are buying from or selling to do not. That “something” could be one or more of many things. Understanding the sector’s tailwinds, potential new products in the pipeline, and new management about changing capital allocation policy. Each boils down to a different and more correct interpretation of the future.

Momentum investing is the opposite. It assumes that, given any company, a set of investors knows better and relies on buying after them.

It logically follows that a momentum strategy never buys at the bottom. Many investors find it hard to buy a stock that is up significantly.

Is wrong, a lot

If we define the “winner ratio” as the percentage of stocks that result in positive returns during their tenure in the portfolio, then Momentum investing barely does 50.

Here’s why that might be. Momentum’s approach is to buy trending stocks. Since a trend is a fad that endures, an approach that looks to get in early sometimes buys fads that die out.

In the long run, half the stocks exit the portfolio lower than they entered (the cluster of red dots in the above chart are live data for Adaptive Momentum). If you’re used to getting most of your questions right on exams at school, this can be a physically painful experience.

The returns in Momentum come from having a few big winners and not losing too much on the losers. Even knowing this in advance doesn’t make it easy to live through.

Casts a wide net

Every stock is not a great investment, but a great investment can come from any stock.

Successful fundamental investors sensibly stick within a circle of competence. Since Momentum has none, it relies on ideas from the thousands of different circles of competence. Its biggest winners could range from old-school commodity manufacturers to new-age internet companies and anywhere in between.

The ability to move to different parts of the market, be it market caps, industries, or companies long ignored by the market, makes Momentum an agile approach. This is also a reason why we are not believers in limiting momentum-factor strategies to specific segments of the market like small-cap only momentum.

Repeatable and Scalable, given Time

Successful authors face the highest pressure when following up on a successful book. Similarly, the traditional investor who sees his portfolio go through a purple patch followed by a lean stretch needs to hang tight or rejig their portfolio to reflect the new reality. There is no easy answer.

Momentum, as a Quantitative approach, relies on staying true to itself by applying a proven strategy that yields outcomes in the long term.

Short-term outcomes are noisy. Long-term outcomes reflect the quality of the process.

Cannot diagnose why it has not been working

When a fundamentally picked stock, like a defence equipment company, does not deliver the expected returns, there is usually an identifiable cause. The orders didn’t come in as expected because the expected spending didn’t materialise. Or the orders came, but the company struggled to deliver due to quality issues, or maybe they delivered but couldn’t convert receivables to cash. Perhaps they did all the above, but the market had priced even higher expectations, leading to a correction.

Momentum strategies go nowhere for periods. For example, in 2018, or for most of 2022. Only in retrospect can we see that most stocks went nowhere during those periods. Not being able to pinpoint specific reasons can be unsatisfying for a lot of investors.

Works because it "stops working"

“My portfolio has now been trailing the benchmark for 12/18/24 months. It has stopped working.” – Momentum investors, just before they throw in the towel

In its 60-month tenure, Adaptive Momentum has trailed the benchmark nearly a third of the time, with one continuous stretch lasting 18 months. Of those 18 months, for 12 months, Adaptive Momentum showed a negative 1-year rolling return. The kicker, this behaviour is almost identical to Momentum backtests.

A fundamental investor can attribute their portfolio struggles to their chosen set of stocks being unpopular and will themselves to be patient. A momentum, or any Quantitative investor, cannot specifically explain why it hasn’t worked for a while, except knowing it doesn’t occasionally work, often for uncomfortably long stretches.

Any strategy with no detractors will stop working because then, who would investors buy from?

Cannot be timed with certainty

This one is the hardest to accept for even long-time practitioners of Momentum investing, us included.

At Capitalmind, we integrated some indicators that work to move into cash when broad market corrections unfold, like the pandemic crash. But knowing the specific periods when Momentum will not work, so far, hasn’t been solved. Conversely, you also can’t forecast when it will start working again.

Backtests and live performance indicate Momentum underperforms the benchmark roughly 1/3^rd of the time on a 1-year basis. When the holding period increases to three and five years, the chances of decent performance increase significantly.

The only workaround, therefore, is to stay consistently allocated.

The best returns from momentum investing come from a combination of a robust underlying momentum investing process and the discipline to keep investing.

Does not beat the "best" fundamental investors

The best investors are contrarians in their consistent ability to buy the stocks where the gap between the market’s assessment of the company’s earning trajectory diverges widely from their own. Consistent is the keyword.

Since Momentum assumes we know less, by definition, it will only buy stocks after the best fundamental investors. So, even the best momentum investing PMS strategy shouldn’t outperform the best fundamental investors.

This means it is not for investors who only accept having “beaten” all other investors as a satisfactory outcome.

But outperforms the ‘Average’ Investor

The average fundamental investor believes in a chosen set of stocks as the holy grail of returns. They identify with their companies and consider all stocks not on their favoured list as inferior investments.

Since it casts a wide net, Momentum Investing has no tribal affiliation with any set of stocks, sector, or theme. The Momentum investor is comfortable changing her mind, so she does better than the average investor.

Momentum’s edge is not what people think

Three underlying edges account for all superior returns, i.e. returns greater than the major benchmark index:

Informational Edge: Having more information than other investors before they do. This includes the illicit kind, where a small set of investors know before the rest. Thanks to capital market regulations, this kind has largely disappeared. The other is doing legwork by visiting channel partners, vendors, customers, and competitors to determine a company’s trajectory before it becomes common knowledge.

Analytical Edge: The ability to better analyse available information. This could involve enhanced skills in assessing markets and securities, proprietary models, and even better workflows.

Behavioural Edge: Capitalises on common investor biases and limitations in decision-making under certainty. It involves structuring a process that avoids emotional biases and consistency pitfalls.

Momentum investors like to believe their edge is analytical. Resting on the sophistication of the model used to score and rank stocks, the frequency at which they rebalance, whether they follow an overlapping portfolio, the position-sizing mechanism and rules that govern moving to cash. Getting those things right is important, but only up to a point.

The Momentum Investor’s edge is primarily behavioural and, to a lesser extent, analytical. The hallmark of that edge is the ability to follow a consistent process without discretionary overrides and stick with it, even when it “feels” painful.

Closing: Momentum, Humans, and Investing

Investment approaches with a solid recent run tend to get the most interest from new investors, often without fully appreciating the nuances. When the inevitable underperformance comes, that initial excitement vaporises, and reallocation decisions get made, only for the cycle to repeat with the next “can’t miss” theme.

The question then is, what are the chances that the collective behaviour that has enabled momentum investing the world over for over two centuries and in the relatively nascent Indian market will cease to exist over the next decade?

• If you’ve walked down a street in an unfamiliar town looking for a meal, chances are you picked the crowded place over the place with empty tables.
• When deciding between two online sellers for identical-looking products, we pick the one with higher ratings and a significantly higher number of reviews.
• Social media algorithms surface the most-watched videos because those are most likely to get the most engagement.
• We are more likely to sign and repost an online petition with thousands of signatures over one with a handful.
• When choosing a hotel for your first trip to a city, we’ll pick a familiar chain over an unknown standalone hotel.

There are logical reasons why we make choices this way. Social Proof is the idea of relying on the wisdom of crowds to filter out undesirable options. The availability heuristic says that popular things that come to mind quickly are reliable. The bandwagon effect says we’re more comfortable climbing aboard a trend when it is widespread.

These tendencies of human behaviour are time-tested. Is it surprising that they extend to investing?

Momentum is not a silver bullet strategy (nothing is) that will magically only buy stocks that go up or that presciently avoid every kind of portfolio decline (because then it’d be more accurately called the Madoff strategy²⁰). Its core strength is in offering a consistent, repeatable way of investing that, however unintuitively, has stood the test of time.

Investing in Momentum is easy. Staying invested in Momentum is harder. Plan accordingly.

---

Get in touch with any questions or comments. I am on Twitter/X @CalmInvestor, email anoop@capitalmind.in. For some of the specifics about Capitalmind Adaptive Momentum, click on More about Adaptive Momentum.  Or schedule a call with a Capitalmind Client Advisor

Notes

¹ Instead of getting lost in a maze of can’t-miss momentum webinars and strategies, we suggest referring to Quantitative Momentum by Wes Gray and Jack Vogel.

² The topic of valuing companies is extensive and the subject of several post-graduate level courses in Finance. Some of the most popular references are ‘Security Analysis’ by Benjamin Graham and David Dodd, ‘Valuation: Measuring and Managing the value of companies’ by McKinsey & Co., and ‘Damodaran on Valuation’ by Prof. Aswath Damodaran.

³ The term was coined by Michael Porter, in his book ‘Competitive Strategy: Techniques for analysing industries and competitors’. Capabilities can be physical, like efficient manufacturing plants and logistics networks, or intangible, like a reputation for great customer service, brands, intellectual property or even access to cheap capital.

⁴ The five tests of a viable quantitative investment thesis: In their book “Your Complete Guide to Factor-Based Investing”, Andrew Berkin and Larry Swedroe list five conditions a quantitative factor should meet.

Applying these criteria filters out a lot of prospective “can’t lose” investment strategies when you see them work for short periods or on a narrow universe of stocks.

⁵ R-squared: An informal explanation: How to interpret R-squared in regression analysis, A more formal one by PennState Eberly College of Science: The Coefficient of Determination, r-squared

⁶ p-value: Pubmed Central - What the p-value really tells us

⁷ Statistical significance: Easy-to-understand article by mailchimp: A Business Owner’s guide to Understanding statistical significance

⁸ The 2012 Forbes article by David Leinweber: Stupid Data Miner Tricks: How Quants fool themselves and the economic indicator in your pants

⁹ ‘Backtesting Portfolios’ presentation by Prof. Daniel Palomar of HKUST

References: Momentum in Global markets and India

¹⁰ Jegadees Narasimhan, Sheridan Titman, Returns to buying winners and selling losers: Implications for stock market efficiency (1993)

¹¹ Tobias Moskowitz, Mark Grinblatt, Do Industries Explain Momentum? (1999)

¹² Yao Hua Ooi, Lasse Heje Pedersen, Tobias Moskowitz, Time Series Momentum (2012)

¹³ Brian Hurst, Tao Hua Ooi, Lasse Heje Pedersen, A century of trend-following investing (2017)

¹⁴ Cliff Asness, Andrea Frazzini, Tobias Moskowitz, Fact, Fiction and Momentum Investing (2014)

¹⁵ Eugene Fama, Kenneth French, Dissecting Anomalies (2008)

¹⁶ Capitalmind, Does Momentum Investing Work in Indian Equities? (2019)

References: Behavioural reasons why Momentum works

¹⁷ Zhi Da, Umit G. Gurun, Mitch Warachka, Frog in the Pan: Continuous Information and Momentum, The Review of Financial Studies, Volume 27, Issue 7, July 2014, Pages 2171–2218

¹⁸ Mark Grinblatt, Bing Han, Prospect theory, Mental Accounting and Momentum, Journal of Financial Economics, Volume 78, 2005

¹⁹ Antoniou, C., Doukas, J., & Subrahmanyam, A. (2013). Cognitive Dissonance, Sentiment, and Momentum. Journal of Financial and Quantitative Analysis, 48(1), 245-275. doi:10.1017/S0022109012000592

²⁰ Bernie Madoff: Who he was, how his Ponzi scheme worked