Towards the end of completing my masters in data science, I started picturing myself doing clever things with machine learning and automated trading. If like me, you have run into the “how do I get historical free tick data” connundrum, then this post is for you. I have structured my post in

Successful and long-lasting quantitative research programs require a solid foundation that includes procurement and curation of data, creation of building blocks for feature engineering, state of the art methodologies, and backtesting. In this project we create a open-source python package

R is an object-oriented programming language and work environment for statistical analysis. It is not just for programmers, but for everyone conducting data analysis, including portfolio managers and traders. Even with limited coding skills R outclasses Excel spreadsheets and boosts information

As an option seller you want the market to stay within the range prognosticated by implied volatility. But what is the historic probability that markets behave as expected? And what other analysis could be done to enhance your chances and find the periods when it is wise to sell an at the money

One of the more popular equity strategies over the past decade is low volatility investing. Simply put, this is a systematic strategy that invests in stocks with lower volatility, either measured by Beta or standard deviation. Why? Well, the low-beta anomaly is the fact that in the past, academics

In my last two posts, Market Timing with a Canary, Gold, Copper, LQD, IEF and much more and Day of Month and Market Timing, I assumed that we earned no interest in cash. Most methods did a good job of telling us when to be in the SPY and when to be in cash. How much could we boost returns by

In a previous post, from way back in August of 2017, we explored the relationship between the VIX and the past, realized volatility of the S&P 500 and reproduced some interesting work from AQR on the meaning of the VIX. With the recent market and VIX rollercoaster, this seemed a good time to

What are the research questions? This paper is the first to examine the impact of including an active share target into the mean-variance optimization process of constructing portfolios. They use the Ceria and Stubbs (2006) approach to robust portfolio optimization as methodology. Monte Carlo

The bond risk premium is the return that investors earn by investing in longer duration bonds. While the most common way that investors can access this return stream is through investing in bond portfolios, bonds often significantly de-risk portfolios and scale back returns. Investors who desire

Given the popularity of Low Volatility, investors might expect structural shifts in the factor characteristics Betas, valuations, sector biases, interest rate sensitivity, and factor exposures are highly time-varying Although these are worth monitoring from a risk perspective, none seem particularly

Are broad measures of public sociopolitical sentiment relevant to investors? Do they predict stock returns as indicators of exuberance and fear? To investigate, we relate S&P 500 Index return and 12-month trailing S&P 500 price-operating earnings ratio (P/E) to the percentage of respondents

The Two Centuries Investments blog from Mikhail Samonov has become a new favorite of mine. More “thought heavy” than “numbers heavy”, Mikhail is a fount of novel ideas. In this piece he describes something I apply in my own investing (though never defined so succinctly): the need to balance

No pain, no premium: risks are the currency that investors need to pay in order to earn excess returns in the long run. Over a period of almost 100 years the Fama-French US market index achieved approximately 10% annualised return but temporarily lost more than 30% on multiple occasions. For no

Moving average is one of the most commonly used smoothing method, basically the go-to. It helps us detect trend in the data by smoothing out short term fluctuations. The computation is trivial: take the most recent k points and simple-average them. Here is how it looks: Moving average example with

This is a summary of the recent performance of a wide range of excellent Tactical Asset Allocation (TAA) strategies, net of transaction costs. These strategies are sourced from books, academic papers, and other publications. While we don’t (yet) include every published TAA model, these strategies

For those of you that don’t know, the idea of creating an open-source package based on Advances in Financial Machine Learning stemmed from Ashu and my master’s project at WorldQuant University. The initial goal was to build out a package that we could leverage to help us extend the literature

My Advisor Perspective article of June 17, 2019 discussed the regime shifting nature of the low volatility anomaly—low volatility stocks have outperformed high volatility stocks, providing both higher returns while experiencing lower volatility. For example, in his 2012 paper, “Enhancing a

Let’s talk about tail risk modelling today. In this blog, I want to introduce Extreme Value Theory (EVT) which concerns itself with modelling of the tails of a distribution, and its key results. As we go along we will work through a toy example with basic R implementation. There are two popular

Alternative Title: What to Do if Your Boss is Terrified of New Ideas? Several younger quants have asked this question: “The culture of our quant group is very skeptical about new ideas. They are terrified of data-mining, and random factors. How can we innovate in such environment?” My thoughts

We have shown many times that timing luck – when a portfolio chooses to rebalance – can have a large impact on the performance of tactical strategies. However, fundamental strategies like value portfolios are susceptible to timing luck, as well. Once the rebalance frequency of a strategy is set,

In financial markets, two of the most common trading strategies used by investors are the momentum and mean reversion strategies. If a stock exhibits momentum (or trending behavior as shown in the figure below), its price on the current period is more likely to increase (decrease) if it has already

For the first time, we will use the features of multiple financial instruments. In this case, we will use the main Forex pairs and the SP500 to perform our forecasts on Gold. This Jupyter notebook will be your guide for more complex calculations. Obviously, you can change the features and

Generative Adversarial Networks are a quite powerful tool for generating synthetic samples. Visual inspection has been used as a traditional measure of performance. However, it is quite hard to inspect when a time series looks realistic or not! Which methodology can be used then? In order to measure

Value and Size have a positive relationship with the PMI, similar to the S&P 500 Indicates that risk sentiment is a core driver of factor performance Investors can consider incorporating variables like the PMI in a risk management framework INTRODUCTION A physicist, a chemist, and an economist

Today’s quantitative investors seem split between innovation on one hand and engineering on the other. The prior group is constantly looking for new factors that predict returns in areas like alternative data and machine learning - yet often fail to find them. The latter camp is focused on

TL;DR: We built a profitable Bitcoin sentiment strategy yielding 2400% returns over 24 months. Adding trading fees made the strategy more realistic while finding optimal sentiment combinations and window sizes increased returns dramatically. In the previous article, we described how to build a

New to Tactical Asset Allocation? Learn more: What is TAA? There are two hard costs that investors must consider when comparing a tactical asset allocation strategy to conventional buy & hold: (1) increased tax liability (if trading in a taxable account), and (2) increased trading costs

Stocks are risky investments. Let's be very clear, stocks are risky with positive skew. Of course, everyone knows that but some data published about two years really drove that home. (See my earlier post "Most stocks are losers - Median and skew tell an important story" about the

In this article I wanted to concentrate on some basic time series analysis, and on efforts to see if there is any simple way we can improve our prediction skills and abilities in order to produce more accurate results. When considering most financial asset price time series you would be forgiven for

We explore a representative multi-asset momentum model that is similar to many bank-based indexes behind structured products and market-linked CDs. With a monthly rebalance cycle, we find substantial timing luck risk. Using the same basic framework, we build a simple ensemble approach, diversifying

Simple value and momentum strategies often end up with opposite market positions. One strategy succeeds when the other fails. There are two plausible reasons for this. First, value investors regularly bet against market trends that appear to ‘have gone too far’ by standard valuation metrics.

The date on which a portfolio is rebalanced can have a tremendous impact on realized performance. We demonstrate that a strategy rebalanced on different dates using the exact same investment process can exhibit return differentials of over 20% across short periods of time. These differences are

A year in the making, my third book is now complete. Trading Evolved is quite different from my previous books, and substantially more information packed. This book is a practical, in-depth guide on how to backtest and analyze strategies using powerful Python techniques. To my knowledge, no such

In this week’s post, we discuss four articles. The size, written by the folks at AQR, is titled “Fact, Fiction, and the Size Effect” and is a deep dive into the Size effect–I highly recommend everyone read the underlying paper as well. The second article examines the baseline historical

As many investors have experienced, Value investing has underperformed for some time now. For the period following the Global Financial Crisis, Value investing (in general) has underperformed (1) the market and (2) Growth stocks. So while the past decade has been rough for Value investors, it can be

Although we find currency risk particularly interesting, it is not often the case with many investors for whom it is no more than a necessary inconvenience. As such, they tend to neglect it, accepting undesirable non-remunerated risks and missing potential opportunities. To prevent this, in this

This is a test of two variations of the Growth-Trend Timing (GTT) strategy from the always thought-provoking Philosophical Economics. GTT combines trends in both price and key economic indicators to switch between US equities and cash. Like most trend-following strategies, the strength of GTT

Because equities are much riskier than high-quality bonds, the vast majority of the risk of a conventional 60 percent equity/40 percent bond portfolio is equity risk. Here’s the simple math demonstrating the point. Well-diversified equity portfolios have volatility of about 20 percent, and

While the SPX closed up the VIX also rose. Most often they trade opposite each other, so this kind of action is somewhat unusual. But VIX has a tendency to decline going into the weekend (Friday afternoons), and then rise when it returns from the weekend. So to see this action on the first trading

In last week’s post, I made a strong assertion that has caused some great feedback and comments. When I first heard Mark Kritzman make a similar point at a UBS conference a few years ago, I had a similar reaction: “Hey, I’m a quant and I love my factors. They are definitely real!”. I still

Part of the reason the market has rallied over the past few days is an indication that a rate cut is likely coming as soon as the next Fed meeting. It is interesting timing for the Fed to begin cutting rates, since their QT program still remains in place (though it is winding down). By reducing the

Investors would like to maximize upside participation while mitigating losses. This preference is at the base of the growth of the liquid insurance market in the form of equity index options. The author investigates the following research question: Are protective put options an effective tail hedge?

In this post, I would like to quickly introduce what I believe to be an underutilized modelling technique that belongs in most analysts’ toolkit: the quantile regression model. As I am discussing some of the main points, I will be working with R’s quantreg package that is maintained by the

Many retirement planning analyses rely on Monte Carlo simulations with static assumptions for withdrawals. Incorporating dynamic spending rules can more closely align the simulations with how investors would likely behave during times when the plan looked like it was on a path to failure. Even a

The ESG factor generated positive returns since 2011 Strong sector biases (long tech & short discretionary) explain the performance Residual returns from ESG investing are essentially zero INTRODUCTION Investing is complicated as it is simple and complex at the same time. Common advice for new

One of the most successful investment strategies since the turn of the century has been the risk-parity “long-long” of combined equity, credit and duration derivatives. In a simple form this trade takes continuous joint equal mark-to-market exposure in equity or credit and duration risk. A

In “Your Complete Guide to Factor-Based Investing,” Andy Berkin and I presented the evidence demonstrating that momentum, both cross-sectional (or relative) momentum and time-series (or absolute, trend following) momentum, not only increases the explanatory power of asset pricing models while

Welcome to the first installment of Reproducible Finance by way of Alpha Architect. For the uninitiated, this series is a bit different than the other stuff on AA – we’ll focus on writing clean, reproducible code, mostly R (but some python too), applied to different ideas from the world of

Yield Curve and Financial Uncertainty: Evidence Based on Us Data Efrem Castelnuovo (University of Melbourne) June 2019 How do short and long term interest rates respond to a jump in financial uncertainty? We address this question by conducting a local projections analysis with US monthly data,

Some price series are mean reverting some of the time, but it is also possible to create portfolios which are specifically constructed to have mean-reverting properties. Series that can be combined to create stationary portfolios are called cointegrating, and there are a bunch of statistical tests