Vance Harwood passed away in September 2026 after a brief and very unexpected illness. Any clients who have posted orders can get their payments refunded via the platform chosen for payment. May his best live on in others. Should you have any important questions or concerns, contact Heidi Nordberg: hlnordberg at gmail

Holidays put people in motion. In the days surrounding major U.S. holidays, airports become busier as travelers visit their families or take advantage of extended weekends. Financial markets themselves are known to display a holiday-related seasonality. In our previous research on the Pre-Holiday Effect in Commodities, we identified a short-term price drift in crude oil and gasoline before major

Every quant hits the same wall. You have hundreds of features that look predictive in isolation. Now you need to combine them into a single portfolio. This is a worked example of that problem: 576 features, 25 instruments, 5 asset classes, daily bars. What to select, how to combine, and where it quietly breaks. 1. The Problem The investment universe is 25 instruments across 5 asset classes: equity

A regime tilt is bought as insurance: lean into momentum while the market is calm, into quality when it is stressed, and keep your head in a crash at little cost in between. The premium is rarely priced with the rule held fixed, because most tests choose the thresholds after seeing the crashes. This one fixed them before the walk and printed the bill year by year. Two portfolios hold thirty

There is 150 years of data on stocks, on bonds, on gold, and on property. On the asset most people actually hold, there is almost nothing. Cash has no volatility, so there seems to be nothing to measure. That turns out to be the wrong conclusion. What is actually guaranteed A Treasury bill promises a number. You put in a hundred, and in three months you get back a hundred plus a little, and the

Trend following is one of the oldest and most persistent anomalies in finance. The evidence that recent winners continue to outperform recent losers has been documented across virtually every liquid asset class, stretching back at least two centuries. It stands in direct opposition to the Efficient Market Hypothesis, yet it has survived out-of-sample testing, multiple market regimes, and decades

The overnight gain is real. Since 1993 nearly everything the S&P 500 paid came between the close and the next morning's open, and a dollar held in SPY only overnight ended more than ten times above a dollar held only through the trading day. Every year someone puts this back into circulation as a discovery, with a table of small stocks whose night returns run to thousands of times and a

A 70.8% continuation rate looks like a trade. Greene sorts 4,061 Kalshi contracts by the quality of their price trend over the window from 30 to 12 minutes before close, and the top decile keeps moving in the trend's direction 70.8% of the time against 51.2% in the bottom decile. Ex-post forecast error falls from 14.75 cents to 6.07 cents across the same sort. Both gaps carry contract-level

When we published that piece we said a re-test was scheduled and that we would report it whichever way it came out. It has now run, on data through 18 September. Here is what it found, and what it changes. The short version. On all the data together the result is stronger and cleaner than what we published. On the five weeks of genuinely new data considered alone, the size of the effect held up

We wanted to see whether a long-only, rules based algorithm applied to US industries could remain profitable over a full century. Our paper written with Gary Antonacci, A Century of Profitable Industry Trends, answers exactly that. Using Kenneth Frenchs industry data from 1926 to 2024, the strategy delivers 18.2% compounded per year, with 12.6% volatility and a Sharpe ratio of 1.39. For

Macroeconomic theory suggests that aggregate demand for goods and services is a key determinant of interest rates. Interest rates regulate demand strength or weakness through market-based financing conditions and central-bank reaction functions. If financial markets do not immediately incorporate all information contained in macroeconomic trends, excess-demand pressures should help predict

Fama and French sort every US stock on the signal at the end of June, hold the top thirty percent weighted by market value, and quote the spread against the bottom thirty percent. A fund that sells the factor holds something near that, restricted to large caps. A private investor who reads about the same factor buys thirty names and weights them equally. The three are assumed to be one trade at

As many of you know, I trade the US, ASX (Australian), and TSX (Canadian) markets systematically. And hopefully in the near future Ill be adding even more markets. It puzzles people why I trade markets other than the US. The US has to be the best, right? Its the biggest and has the most volume, so surely every trading opportunity I could ever need lives in the US market… right? I get this

The first two autopsies of this series documented two species of the same family of errors: the inclusion look-ahead, which contaminates the list of names that enter the backtest, and the execution look-ahead, which contaminates the price at which the purchase is assumed to have happened. The present note documents the third species, found in an experiment on shareholder activism, and with it the

A large-scale empirical test of whether within-strategy Monte Carlo permutation testing, a standard validation step in quantitative strategy development, actually improves forward-looking strategy selection. Across 437,911 strategy configurations on nine instruments spanning crypto, forex and commodities, 160 walk-forward windows, and 26.5 billion permutations end-to-end, the answer is no, but for

Bloomberg started selling the clock inside the bar. Not just the high price for the minute, but the second at which the high printed, the same for the low, the first and last trade times, condition-code tick counts, and a Bloomberg VWAP for the interval. Ruslan Tepelyan trained three neural networks on that feed across the Russell 3000 for all of 2021 and reports that the timing features improve

Welcome to 2026, where bonds no longer diversify stocks, volatility remains elevated and inflation isnt going anywhere1. This is the dramatic, although factual, opening of Basis Pointss video Morgan Stanleys asset allocation playbook for the next 5 years. As a consequence, the 60/40 Portfolio – invested 60% in stocks and 40% in bonds – which for decades has stood as the institutional

Every trading day brings a wave of upgrades and downgrades, each one moving a share price and none of them ever audited. The reason is mundane: to score a call you need the price on the day it was made, and that is the one thing almost nobody keeps. We keep it. Our records hold 149,587 analyst rating changes with the price at the moment of the call, going back to May 2016, with the firm named on

Fit a connected piecewise-linear trend to the Swiss Leaders Index with fifteen breakpoints and you get an in-sample R-squared of 0.954. Fit the same model to a driftless random walk of the same length and you get 0.966. I ran 200 of them. The noise wins at every single value of K, from zero breakpoints to fifteen. Whatever the 0.954 is measuring, it is not evidence that the SLI has regimes. This

The explosive growth of hyper-liquid 0-DTE markets has pushed traditional options pricing infrastructure to its breaking point, as continuous Black-Scholes calculus can collapse into an unusable point mass at expiration. Rather than patching a broken formula with hand-fitted tweaks, a new paper suggests dismantling legacy math by replacing continuous geometric Brownian motion with a discrete,

Reflexivity is the process through which market participants actions influence the very market variables on which their decisions are based, creating feedback effects. In this post, we discuss how reflexivity manifests itself in options hedging and how it can be explicitly incorporated into option pricing models. Feedback Effect in the Foreign Exchange Market The BlackScholes-Merton (BSM)

Piotroski's F-score is the screen every value investor has run: nine yes-or-no questions on the last two annual statements, a point for each yes, buy the eights and nines. Piotroski (2000) built it for the cheapest fifth of the market by book-to-market, where a positive return on assets was information and the high scorers earned 7.5 percentage points a year more than the group between 1976

This report evaluates the Breakout strategy through signal diagnostics, parameter selection, robustness testing, and portfolio integration. It examines whether historical performance holds up under recent conditions and whether the strategy adds value to the portfolio. Despite strong results over the full sample, weaker recent performance and substantial overlap with existing strategies support

Dividend capture is sold as income on a calendar: buy the stock at the close before the ex-date, the payment is yours, sell the next day. On the ex-date the price falls, and the trader keeps only the part of the payment the price did not take back, minus the spread both ways. Elton and Gruber measured the fall at about four fifths of the payment in 1970. We expected the fall to match the payment

We track 100+ Tactical Asset Allocation (TAA) strategies. A unique feature of our platform is our Aggregate Allocation Report, a daily snapshot of the average asset allocation across all of the strategies we track. To illustrate, in the graph below we show the aggregate allocation over the last 3 years, summarized by asset category. Note the increase in defensive allocation (ex. cash and bonds)

The first two articles in this series measured what one countrys stocks and bonds did over 150 years and found that the country decided most of it. This one holds all sixteen, adds gold, and asks what the combination does in each economic regime, and whether the trend rule from the fourth trend-following article changes the answer. What is held The equity sleeve holds sixteen countries in equal

Every European rotation system in circulation treats the country as the unit of risk. The people who run these books rank Germany against Spain against Sweden and call the result a country bet. A European country index is a concentrated sector bet: banks carry the south, pharma and staples carry Switzerland. So the question here is whether European country rotation is already sector rotation, and

As we recently anticipated to our readers, we are expanding our research effort into listed US options. As usual, the first step is building a reliable historical database. For our work on SPX options, that means finding a source of clean, consistent, and usable intraday data. Cboe DataShop is widely used in both academic and practitioner research on SPX options. Several recent studies on 0DTE

Well yes, I am pleased to announce that as of this week I completed final proof reading of my new book "The Art And Science of Trading"(AAST). It joins the list of my existing books with their own acronyms: Systematic Trading (ST), Smart Portfolios (SP), Leveraged Trading (LT) and Advanced Futures Trading Strategies (AFTS). Due to the inherent delays involved in global dead tree supply

The previous note showed that a backtest with a net Sharpe of 0.63 fell to 0.06 upon applying a single filter: the point-in-time membership of the index, that is, knowing with precision which companies belonged to the S&P 500 in each month of the sample. It should be noted that a paradox was then only hinted at: the most decisive input behind that result was not any price, nor any volume

Nine cohorts are cut at every anchor from the point-in-time S&P 500, three terciles of earnings predictability crossed with three of gross profitability, and each walks twenty sealed one-year windows from 2006 to 2025 at four holding periods in two book shapes: a long-short book that buys qualifying beats and shorts qualifying misses, and a long-only pair, every announcement of the cohort

The integration of artificial intelligence into algorithmic trading has ignited a race to transform generative text into systematic alpha. A new paper written by Steven Edwards empirically investigates whether constructing a synthetic consensus using large language models can simulate information aggregation dynamics or if it merely acts as a sophisticated echo chamber. By utilizing an expansive

The drift the academic literature describes does not pay in large caps: form the classic quarterly surprise book and the grid nets -0.5 percent a year across eleven sectors and twenty years. The reaction to the earnings number itself still does: enter each company the day its own number is known and the same surprise measure nets +2.2 percent a year, peaks at the ten-day hold, and fades by the

Over the past year I built an operating system for quantitative trading research: a harness that lets AI agents run the grunt work of strategy development end to end, inside rails that make self-deception structurally difficult. It recently carried two strategies from an empty folder through the full gauntlet: one crypto trend follower to a deployable, drawdown-controlled config, and one options

Trends in gold returns can be partly explained by the macroeconomic environment. Persistent monetary easing, dollar stability risks, and weak economic sentiment can each drive sustained demand for gold. This article shows how to construct simple point-in-time macro factors that capture these themes and combine them into a broad macro-support score that can serve as a trading signal for

One of the most enduring puzzles in finance is the apparent disconnect between Wall Street and Main Streetmarkets sometimes soar while the underlying economy stumbles, and vice versa. Paul Samuelson famously quipped that the stock market has predicted nine out of the last five recessions capturing the frustration economists and investors have long felt trying to extract reliable

Beta began as a way to compress a covariance matrix. The CAPM turned it into the equilibrium measure of exposure to aggregate market risk, priced by the market premium. Markowitz maps beliefs to portfolios. With common beliefs, a risk-free asset and market clearing, the CAPM adds one decisive identity: the common tangency portfolio must be the value-weighted market portfolio. The Capital Market

The literature on crypto quant trading keeps citing cross-sectional momentum as a durable edge with Sharpe around 1.1 to 1.5. I built exactly the paper spec, ran it against 303 weekly rebalances across 10 majors, no parameter tuning, no regime filters, no cheating. Annualized Sharpe came out at 0.37. It has been negative for the last 86 weeks straight. Here's what the run actually looked like

For the past months, almost every free hour I had went into one thing. Today its live. eigenscore.com In quant, everyone claims to be good, and there's no arena to settle it. Competitive programming solved that twenty years ago with rated contests. I built the same thing for us. What it is EigenScore is a competitive platform for quants: a problem archive and rated contests with a

The most liquid fund shelf a person can buy holds 3.3 independent bets. Dalio's chart needs fifteen. The Holy Grail of investing, in his words: find fifteen good, uncorrelated return streams and risk falls by roughly eighty percent while return stands still. At correlation zero, fifteen equal streams carry a quarter of one stream's volatility. Whether the shelf a person can actually buy

For those unaware, you generally make more money overnight than you do during the day when holding stocks. So much so that even if you just held the SPY overnight (buy at close, sell at market open the next day), you would net considerably more than if you held it during the day (buy at open, sell at close) Chart, line chart Description automatically generated This is known as the Overnight

I gave a research agent four years of prices with no predictable structure in them none, by construction and it came back with a long/short book, an in-sample Sharpe of 2.1, and a paragraph explaining the economics of an effect that does not exist. That is the measurement in this post. The more useful result is the second one: 88% of that number is accounted for by two integers how

People usually take All Weather for the fixed allocation Dalio gave Tony Robbins, which is the one every retail article prints, but that one is All Seasons and it carries no borrowing at all. All Weather proper is risk parity borrowed up until the book carries a 60/40's volatility, which means the borrowing is the strategy itself. We built that one and charged it the three-month Treasury bill

A 50/50 stock-bond portfolio that earned its target return with moderate volatility before 2000 requires an 80/20 allocation after 2000 to maintain the same profile. That 30-point shift is not a style choice or a bet on equities. It is the portfolio adjustment needed to compensate for the stock-bond correlation flipping from positive to negative, which McMillan documents across the G7 using

Twice in a single afternoon, it told me an idea was good. Twice, it was wrong, and the thing that caught the error both times was a number the headline verdict had quietly averaged away. Heres what the aggregate hid, and why I now trust the breakdown over the bottom line every time. The idea that survived The first idea was a zone-cycle pattern, price making a particular round-trip

Bitcoin's volatility does not mix randomly across time. High-vol periods follow high-vol periods, low-vol stretches extend themselves, and the transition probabilities are stable enough to build a forecast from. Borrego Roldn's 2024 study quantifies this with hourly and daily data from August 2018 to November 2024, constructing 3-state Markov chains around rolling log-volatility and

Most Tactical Asset Allocation (TAA) strategies trade once per month. Thats by design. Short-term market movement is mostly noise, and trying to time every zig and zag is a fools errand. A unique feature of our platform is the ability to follow these monthly strategies on any day of the month. Were not just executing the same signal on a different date were recalculating the signal

Turkish savers have argued about the same question for decades: keep the money in a lira time deposit and collect the interest, or convert it to dollars and hold. The dilemma is common enough to have a household name faiz mi, dolar m?, interest or dollars? Both answers have had long stretches of looking right, which is why the argument never ends. Figure 1 settles it the only way it

Financial markets process an enormous volume of corporate news every day. Earnings announcements, management changes, clinical trial results, insider transactions, partnerships, and regulatory developments can all affect a companys value. The challenge is not simply identifying whether a headline sounds positive or negative. Investors must understand its economic implications, anticipate how

Time-series momentum stands as one of the most reliable and heavily backtested anomalies in quantitative finance, serving as a foundational alpha source for modern managed futures and trend-following strategies. However, a recent academic paper by Matti Suominen and Erik Hjalmarsson, titled Boundaries of Time Series Momentum, uncovers a structural vulnerability that every practitioner must

In this episode of House of Quants, listeners will discover: Why quantitative trading is engineering: The episode challenges the myth of the lone trader searching for magical chart patterns and explains how institutional research operates as an industrial-scale system built around data, statistics, software, execution, and risk. How ideas become executable trading systems: Every strategy passes

Joel Greenblatt's Magic Formula ranks companies on two numbers and buys the ones that score well on both. One asks whether a business is cheap. The other asks whether it is any good. The claim in the book is that the pair works better than either number on its own. We tested that claim by running all three books over the same twenty one-year windows of the S&P 500, from 2006 to 2025.

The obvious rule is to buy what members of Congress buy and avoid, or short, what they sell. Across 17,859 investable common-stock disclosure events, that rule fails. One year after a disclosure became public, the typical purchase event trailed the S&P 500 by 5.36 percentage points and the typical sale event trailed by 6.39 points. Purchases did only 1.13 points better than sales on matched

Easley, O'Hara, Yang and Zhang take five textbook microstructure measures, compute them on Binance one-minute bars for BTC, ETH, XRP, SOL and ADA over January 2021 to July 2023, and ask a random forest to call the sign of tomorrow's change in the return distribution. Two numbers carry the paper. Crypto VPIN averages 0.47 against the 0.22 that the same authors found in E-mini and crude

Markowitz's 1952 paper deliberately begins after investors have formed beliefs about future returns. Portfolio choice is the second stage; estimating its inputs is the first. Rebuilding Robert Shiller's three-asset classroom example with two defensible US equity series moves the recommended equity weight from 38% to 26%, while estimated portfolio risk and return barely move. Repeating