Quant MashupEmail summaries are back. Get the Quant Mashup delivered to your inbox every 1-3 days. Never more than one email per day. No spam and no sharing your address. Look for the email link in the header and sidebar. - Mike The Questions That Change Everything [Robot Wealth]Simon from The Algorithmic Advantage had me on the show, and the episode went up last week. We hadn’t prepared much, which turned out to be the right call, because what came out is about the most complete walk-through of how I think about trading that exists in one place. Seventy-two minutes, from(...) How Market Feedback Loops Affect Alpha [Delphic Alpha]Momentum is usually profitable on strong trends. The literature on feedback trading and liquidity predicts one exception: the week after a trend moved further while liquidity was low. I combine four strands of that literature into one model and test it on daily data for ETFs, futures, S&P 500(...) Quant Observer: A curated directory of quant tools, data, research, etc.quant.observer collects useful quant trading tools and frontier research in one place. The goal is to save time finding reliable infrastructure and reading material. We do not run a trading platform or offer investment advice. Everything here is reference material for your own work. The insider-buying edge was mostly 2008 [Donttrade]When several company insiders buy their own stock in the same week, the stock tends to outperform. It is one of the better-documented anomalies in finance, it has a mechanism you can state in a sentence — the people who know most about a business are putting their own money into it — and it has(...) TTM Squeeze and Bollinger Squeeze: breakout or volatility clustering? [Quanter Lab]Chart books call the squeeze the calm before the storm: when a stock's Bollinger Bands slip inside its Keltner Channel, a big move is supposed to be near. Econometricians say calm is followed by more calm, and volatility by more volatility. We measured both on S&P 500 stocks from 2006 to(...) From Signal to Bet: A Blueprint for Multi-Asset, Multi-Horizon Trading Pipelines [Delphic Alpha]A system that runs several models across many assets, bar sizes and forecast horizons has to answer the same questions for every prediction. How long is the call held? How big is the bet? Which calls are traded at all? How do positions from models on different clocks become one position per symbol?(...) Does Empirical Finance Replicate? AI answers the question. [Alpha Architect]Published research often becomes part of the permanent investment toolkit. Findings are cited, models are calibrated, and strategies are built. Empirical relationships estimated decades ago can continue to influence decisions long after the original sample ends, but there are really two different(...) Why Long Bonds Still Belong in the Portfolio [Beyond Passive]In the first week of October the thirty-year Treasury yield closed at 5.60%, its highest since 2002. TLT, the long-bond fund in the three-asset, risk-balanced portfolio this publication uses as its baseline for a long-term macro strategy, trades at a price it last saw twenty-two years ago. Ray Dalio(...) Momentum Mini-Portfolio Development - Part 4: ETF Diversifier [TradeQuantiX]At the end of Part 3 I said I wasn’t exactly sure where this series would go next... But after some brainstorming, the answer was pretty clear. We already built a TSX momentum system, a US momentum system and an ASX momentum system. Three different markets and three different ways of trading(...) After a Strong Pre-Open Hour the Move After the Open Is Bigger, Not Smaller [Krueger Algorithms]A common expectation: if the market has run hard in the hour before the cash open (start of the regular session), part of the move is done and less comes at the open. We ask about the size of the move at the open, not its direction. The push is the close of the last minute before the open minus the(...) Do markets hunt the stops beyond yesterday's high and low? [Daru Finance]Trading forums, indicator libraries and paid courses repeat a claim often enough that most traders take it as given: stop orders pile up just beyond obvious levels such as yesterday's high or low, larger players push price through to fill against them, after which price reverses. Traders act on(...) Podcast with Kris Longmore of Robot Wealth [Algorithmic Advantage]Ex-prop trader on why solo trading is a different game, and how to win with real edges & portfolios of noisy strategies Rather than writing an article along with this pod, I think the relevant piece to refer you to is the series I’m currently writing called ‘The Part Time Trader’ because I(...) Trading the Multi-Asset Drift Around U.S. Elections [Quantpedia]We analyze a multi-asset calendar anomaly around U.S. federal elections that is consistent across assets and simple to trade. While our previous study documents a robust pre-election drift, the cross-asset pricing of its D+1 resolution remains unmapped beyond SPY. Using a diversified ETF basket(...) The Market Reads the Announcement. By the Time It’s News, It’s Old. [Tommi Johnsen]For three months we thought we had found that good news pushes a stock up the following day. Here is what was really going on. When a company reports its quarterly earnings, it often does so just after the stock market closes at 4 pm. Traders read the announcement within minutes and buy or sell in(...) Frog-in-the-pan momentum tested on the S&P 500 [Quanter Lab]Momentum investors buy last year's winners. A well-known refinement, the frog in the pan of Da, Gurun and Warachka (2014), also looks at how a winner went up. A stock that climbed a little on most days is said to be a better buy than one that rose in a few big jumps, because news that arrives(...) A Century of Meme Stocks and the Price of Coordination [Quantpedia]Meme stocks are commonly studied through social-media activity, but this approach limits both the historical scope of research and the signals available to practitioners. In A Century of Meme Stocks and the Price of Coordination, Chad Schmerling develops an alternative: a machine-learning model(...) Building a Statistically Valid Backtest: The Checklist Nobody Runs [Aligrithm]McLean and Pontiff record stock-portfolio predictors whose performance is 26% lower out of sample than in sample, and 58% lower after publication. Suhonen, Lennkh, and Perez look at 215 smart-beta strategies banks offered and find a median deterioration of 73% in the Sharpe ratio from the backtest(...) What Happens After an Analyst Downgrade? [Talval]In August we scored 22,000 analyst upgrades and downgrades against the market and found that the median one is a coin flip. That was the average call. This study asks what the average hides: does the same rating change mean something different depending on what the stock had already done? It does,(...) The 200-day moving average vs volatility targeting [Quanter Lab]Anyone who held stocks through 2008 or the Covid crash has asked the same question: is there a simple rule that gets you out before the worst of it? Two answers are famous. The 200-day line sells when the price ends the day below its average of the last 200 trading days. Volatility targeting holds(...) Orthogonal Cluster Risk Parity (OCRP) [CSS Analytics]luster Risk Parity (CRP)was originally introduced in 2013 and conceived in 2012 by David Varadi and Michael Kapler (Systematic Investor). It was designed as an attempt to make ERC by Maillard, and Roncalli aware of the portfolio’s total asset allocation and allocate risk contributions equally both(...) Buffett's be greedy when others are fearful, tested [Quanter Lab]"Be greedy when others are fearful" is Warren Buffett's best-known advice, and CNN's Fear & Greed Index is how many people check what the others feel. We rebuilt CNN's index from the seven market measures it combines, which takes it back to 2007, and followed the advice(...) Polynomial Regression Bands With a Profit Factor Below One [Aligrithm]Gil Cohen fits second-, third-, and fourth-degree polynomial moving regression bands to Nasdaq-100 names from 2017 through March 2024, charges 0.3% on each fill, and names the four-degree model best. Table 3 puts the average net profit at 162.75 dollars per name. The abstract, the conclusion, and(...) What Really Drives the Asset Growth Anomaly? New Evidence Points to Mispricing, Not Risk [Alpha Architect]One of the most well-documented patterns in the cross-section of stock returns is that firms with high asset growth subsequently underperform firms with low asset growth. This “asset growth anomaly” is so well established that it now sits at the core of two of the most widely used benchmark(...) The turn of the month effect tested: where the stock market's payday is now [Quanter Lab]For most of the last century US stocks made their money on four days a month, the last trading day of the month and the first three of the next, which economists call the turn of the month. The usual explanation is a payday, because salaries and pensions arrive at the end of the month and part of(...) Practical transaction cost checks for macro trading strategies [Macrosynergy]This article demonstrates practical transaction-cost checks for systematic macro trading strategies, using a Python class from the Macrosynergy package. The method requires only basic estimates of transaction sizes and costs, which can often be obtained from trading desks or through LLM queries.(...) 897,000 Tests, About 150 Pattern Families, One Confirmed Finding [Krueger Algorithms]The question was simple: what do you find if you forget everything you think you know and search the data from scratch? No own setups, no favourite ideas, no weighting by gut feeling. Instead, every pattern family that books, forums and papers call profitable, on the same data and under the same(...) How Commodity Prices Affect Stocks: 13 Oil Spikes Since 1970 [Portfolio Terminal]US crude oil (WTI) rose 50% or more within a year 13 times from 1970 to August 2026, and US CPI inflation was higher at every one of the 13 oil peaks than a year earlier, by a median 2.0 percentage points. In the 8 oil spikes since 1970 that took WTI crude to a 5-year high, the S&P 500 fell in 5(...) What Held Up When Bond Yields Rose? 63 Years of Data [Portfolio Terminal]In the 7 episodes from 1963 to 2026 when the 10-year Treasury yield rose 1 point or more in a year and the S&P 500 fell, 10-year Treasuries lost money all 7 times (median −5.5%) and 1-month Treasury bills gained all 7 times (median +6.1%). In those 7 episodes a 60/40 portfolio of US stocks and(...) Is Attention a Factor? WSB Herding, BERT Sentiment, and the Meme Confound [Aligrithm]Huang and Shum Nolan buy three names. Each month they take the tickers WallStreetBets mentioned most, keep a name only when the posts were majority bullish, and hold the equal-weight book for the next month. Regress that book's daily percent return on the market, size, value, and momentum(...) A New Stage, a New Deadline: Quantpedia Awards 2027 Are Here Again! [Quantpedia]Hello everyone, The Quantpedia Awards are back—and this time, we’re taking the winners’ announcement to the stage! For the 2027 edition, we’re bringing together an attractive prize pool, a panel of investment professionals and academics, and a new opportunity to put outstanding quantitative(...) Replication: Intraday Momentum, Eight Years Later [Dead Signals Lab]The previous note inaugurated the replication arc with an anomaly that died by decay. The present replication examines a more recent published case of finer mechanics: the market intraday momentum documented by Gao, Han, Li and Zhou in 2018, according to which the sign of the first half hour of the(...) ML in the Cross Section: Avramov's Companion to the DDA3600 Spine [Aligrithm]On the non-microcap book, 0.50% times the Gu-Kelly-Xiu neural net's turnover of 0.869 costs 0.4345% a month. Its Fama-French six-factor alpha on that book is 0.312%. The ticket is larger than the alpha. Instrumented principal components, the linear model that lets betas move with firm(...) Analysing 335 Quant Trading Podcasts: How Systematic Managers Trade [Delphic Alpha]What do systematic managers actually do with their money? This post distils 335 podcast episodes, about 3.7 million words of interviews with CTAs, quants and allocators, into practical lessons. Every claim links to the episode it came from. It is organised around the questions people ask most: which(...) Correlation, Volatility-of-Volatility, and Sector Implied Volatility [Relative Value Arbitrage]Correlation is an important component of portfolio and risk management. However, unlike volatility, which has received significant attention and for which numerous models have been developed, correlations have received considerably less attention from a modeling perspective. In this post, we give(...) Do Stocks Fall When Bond Yields Rise? 63 Years of Data [Portfolio Terminal]From January 1963 to August 2026, the 10-year Treasury yield rose 1 percentage point or more within 12 months in 20 separate episodes; the S&P 500 fell over the same 12 months in 7 of them. All 7 of those S&P 500 losses came with US consumer-price inflation at 3% or more. In the 6 episodes(...) Information Theory for Traders: Entropy, Mutual Information, Channel Limits [Aligrithm]A feature that matches the sign of the next E-mini return on 4 days out of 5 leaves 0.722 bits of uncertainty about that sign. The part the feature removes is 0.278 bits of mutual information, out of the 1 bit in a fair coin. The binary channel that flips 1 transmitted bit in 5 has that 0.278 as its(...) Kalman filter vs the 200-day moving average: a stochastic model against a geometric rule [Quanter Lab]Many trend rules can be drawn on the chart, and the 200-day line is the best known of them: hold the fund while the price is above its average of the last 200 days, step aside when it falls below. A Kalman filter comes at the same prices from statistics. It treats every close as a noisy reading of a(...) Lead-Lag Is Multidimensional: Trade Price and OBI Predict Others' Midpoints [Aligrithm]The Euro Stoxx 50 future's trade price led the DAX future's midpoint on 99% of trading days from January to June 2021, and the lag that maximized the correlation was 100 microseconds. Bender, Cestonaro, and Schmidt measure that lead on 19 Xetra and Eurex instruments, across nine(...) Can Weakening Morning Order Flow Predict SPY Reversals? [Quantpedia]In a previous article Building and Testing Trend-Following Strategies on One-Minute SPY Data, we investigated whether retail activity indicators derived from one-minute SPY data could be used to construct profitable trend-following strategies. The results suggested that we are able to construct(...) The NAAIM Exposure Index: Contrarian or Continuation Indicator? [Portfolio Optimizer]In a previous blog post, I described the NAAIM Exposure Index, which represents the average exposure to U.S. equity markets as reported by members of the National Association of Active Investment Managers (NAAIM) in a weekly survey. At the end of August 2026, the NAAIM Exposure Index hits a value of(...) RSI(2) mean reversion on the S&P 500, 2006 to 2025: dips, trading costs and the VIX [Quanter Lab]Mean reversion is the idea that a price pushed too far in a few days tends to come part of the way back. The research explains it as a trade: holders who must sell at once push the price below what the news justifies, and whoever buys from them is paid when it drifts back. We test its plainest(...) Stat-Arb Is a Clustering Problem: Multi-View Spectral > Any Signal [Aligrithm]On S&P 500 names from 2000 to 2022, the best book in Raymond Leung's grid is multi-view co-regularized spectral clustering with 25 clusters, a plain long-short spread, and an Ornstein-Uhlenbeck entry and exit. After a 5 basis point charge on each leg at entry and again at exit, that book(...) Stop Using Pairwise Granger (causality upgrade) [Aligrithm]A pairwise screen of the 64-futures book, at a 5% level and five lags, expects 1,008 false links when every cross-link is null. That book is the one in the old article "Network Momentum as a Cross-Asset Factor," and the screen is the bivariate test the old article "Stop Using Pairwise(...) Tech companies that spend more on capex than their operating cash flow [Quanter Lab]Before the dot-com crash, the telecom companies building the internet's networks spent more on buildings and equipment than their businesses brought in, and borrowed or sold shares to pay the difference. Our thirty-year study of the tech industry found that their spending passed their own cash(...) Which Conformal Method Works for Which Alpha Signal? [Delphic Alpha]Conformal prediction is not one technique - it is a family of methods, each with its own strengths and weaknesses. Split Conformal, Mondrian, ACI, CQR - they all produce valid prediction intervals, but they behave very differently depending on the signal you feed them. It is not an implementation(...) Moving Average Distance: The Technical Indicator That Passed the Cross-Section [Aligrithm]Divide a stock's 21-day moving average by its 200-day moving average. That is the whole signal, and it is about as retail as a signal gets. Avramov, Kaplanski and Subrahmanyam ran it across 13,828 US firms and 1,353,679 monthly returns from July 1977 through December 2018, and the(...) The Part Time Trader Part 2: The Questions Before the Backtest [Algorithmic Advantage]Part 1 was survival first: start slow, start boring, and treat the early years as skill acquisition rather than a race to 100% a year. This part is the bridge from that idea into the actual research. The subject is relative momentum: rank a universe of stocks by how strongly they’ve risen, hold(...) Building and Testing Trend-Following Strategies on One-Minute SPY Data [Quantpedia]Intraday trading strategies have gained increasing attention as advances in computing power and market data availability have made intraday strategy analysis more accessible. While many trading strategies are traditionally developed and evaluated using daily price data, shorter timeframes can(...) I Tried A Foundational Financial LLM Model to See If It Holds Up to It’s Claims: It Doesn’t [Paper to Profit]Since the dawn of the LLM age (only few years ago), there has been a flurry of alternative approaches and fundamental remixed on the classic ‘ChatGPT’ style attention transformer. In 2023, Bloomberg released BloombertGPT which was the first ‘financial LLM’ and claimed that it could(...) State-Dependent (In)Efficiency: Meta-Learning a Directional-Change Threshold [Aligrithm]Barak, Razmi and Mousavi report an out-of-sample Sharpe ratio of 1.34 against 0.59 for the best static version of the identical trading logic, on 50 crypto futures from January 2022 to January 2024, with a Ledoit-Wolf bootstrap p-value of 0.008 on the difference. The machine doing the work is a(...)