Building CetaResearch. Data analysis for policy, research, and startups.
pairs trading
Most trading strategies bet on direction. Bull or bear, up or down. Pairs trading takes a different approach. It bets that two related stocks will snap back together after temporarily diverging. The market can go anywhere. What matters is the relationship between two specific stocks. Contents 1. Method 2. What
pairs trading
We screened 3,701 US large-cap stocks down to 2,579 highly correlated candidate pairs. Then we ran statistical cointegration tests on all of them. Only 516 passed (20.0%). Here's what that means, why it matters, and what the 80% failure rate tells you about pairs trading
pairs trading
Finding Pairs Trading Candidates: What 884,000 Correlation Tests Reveal There are roughly 3,700 US stocks with market cap above $1 billion. That's about 6.8 million possible pairs. Most have no useful relationship. Contents 1. The Screening Pipeline 2. Step 1: Define the Universe 3. Step
backtests
Graham Number Backtest: 25 Years of Buying Stocks Below Intrinsic Value Benjamin Graham's formula is simple: multiply EPS by book value per share, multiply that by 22.5, take the square root. If the stock trades below that number, it's cheap. We ran this screen across
analyst ratings
We measured 136,610 analyst rating revision events, upgrades and downgrades, across US, Germany, UK, Switzerland, and Canada from 2012 to 2025. Every market shows immediate price reaction on the day of the upgrade. What happens over the following three months is where markets diverge. Contents 1. Method 2. The
analyst ratings
Analyst Upgrade Clusters Across 9 Markets: A Data Quality Honest Assessment Upgrade clusters, events where 2+ analysts move bullish simultaneously, produce statistically significant next-day abnormal returns in 7 of the 9 markets we tested. But the global story is more complicated than the headline suggests, and understanding why requires looking
analyst ratings
Analyst Upgrades on UK Stocks: Sustained Drift and a Strong Cluster Effect We measured 9,059 individual analyst upgrades on LSE-listed UK stocks from 2012 to 2025. Upgrades produce immediate price impact and the drift sustains through three months. The cluster effect, multiple independent firms upgrading within 30 days, produces
analyst ratings
We ran an event study on 13 years of individual analyst rating changes on US stocks. The average upgrade produces +0.65% abnormal return on day one, but not all upgrades are equal: when two or more analysts upgrade the same stock within 30 days of each other, the drift
analyst ratings
Analyst Upgrades in Germany: Drift Grows to +2.6% Over 3 Months We measured 12,585 individual analyst upgrades on XETRA-listed German stocks from 2012 to 2025 and found something different from the US: the drift doesn't fade. It keeps growing for three months. Contents 1. Method 2.
analyst ratings
When one analyst upgrades a stock, the market shrugs. When three analysts upgrade the same stock in the same month, something measurable happens. We tested this across NYSE, NASDAQ, and AMEX from 2019 to 2025 using aggregate rating count data, and found statistically significant abnormal returns at every window out
piotroski
An accounting professor created a nine-point checklist to separate winning value stocks from future bankruptcies. We ran it on US equities from 1985 to 2025. The high-quality basket (Score 8-9) returned 10.6% annually with a Sharpe of 0.297. The low-quality basket (Score 0-2) returned 10.9% with a
piotroski
We'll be direct: the Piotroski F-Score does not work in India. Over 29 years on BSE and NSE, high-scoring value stocks (8-9) returned just 1.9% annually. Low-scoring stocks (0-2) returned 8.6%. The spread is negative: -6.7%. Contents 1. Method 2. What We Found 3. Alpha