How does the evolving set of artificial intelligence (AI) platforms based on large language models view the likely winners of the AI platform competition. To investigate, we pose to each of Grok, ChatGPT, Claude, Perplexity and Gemini the following prompt: Based on up-to-date market data, please concisely estimate which one or two AI platforms are most likely to achieve and…
Evidence-based investing research since 2004
Research summaries, analyses and reviews for serious investors, financial advisors and money managers: about 20 new studies a month, 3,501 in the archive, and two ETF strategies with monthly signals.
Read 3 premium articles a month, free
No password or card. Or join CXO Premium for everything, $17.99 a month.
Recent Investing Research
Technical Trading Rules for GLD?
Do artificial intelligence (AI) platforms based on large language models offer a quick way to identify potentially exploitable technical trading rules in financial series? As a simple exploration, we pose to Claude the following prompt: Please examine the attached daily…
Lessons Learned from Individual Investor Use of AIs
What lessons are emerging from individual investor use of artificial intelligence (AI) platforms based on large language models (LLM) as an aid? As a simple exploration, we ask each of Grok, ChatGPT, Claude, Perplexity and Gemini to search for examples…
Human Interactions with AIs re Financial Advice
How should investors view financial advice from artificial intelligence (AI) models based on large language models, which have neither finance credentials nor investing track records? In their October 2026 paper entitled "Judge for Yourself: Confidence Beats Accuracy in AI Financial…
Using the AI Panel to Evaluate an Active ETF
Does the evolving set of artificial intelligence (AI) platforms based on large language models offer a quick way to evaluate an active exchange-traded fund (ETF) that "seeks to offer a distinct source of monthly income by selling a 30-barrier put…
Using AI to Summarize/Check Academic Research
In their October 2026 paper entitled "When Price Charts Rhyme: Historical Visual Retrieval for Stock Return Prediction", Hui Chen, Haoyu Liu, Ke Wu and Dexin Zhou report gross annualized hedge portfolio returns as high as 85% and gross annualized Sharpe…
Aggregate Net Stock Buybacks and Future Stock Market Returns
Is aggregate net U.S. public company stock buyback activity a useful predictor of U.S. stock market return? To investigate, we: Retrieve Nonfinancial Corporate Business; Corporate Equities; Liability, Transactions and multiply values by -1 as a quarterly aggregate net stock buyback…
Turn of the Year and Size in U.S. Equities
Is there a material and reliable market capitalization (size) effect among U.S. stocks around the turn-of-the-year (TOTY)? To check, we track cumulative returns from 20 trading days before through 20 trading days after the end of the calendar year for…
Giving Claude More Freedom to Compete with SACEMs
Claude's best-CAGR (11.9%) and best-Sharpe (11.6% CAGR) picks from 480 variants trailed SACEMS EW Top 2 and Top 3 (12.6% and 13.1% gross CAGR; equal-weighted top-N asset class ETFs by momentum; Feb 2007–Sep 2026; in-sample).
Asking Claude to Compete (in Limited Fashion) with SACEMs
Claude's EW Top 4 6-month momentum strategy (monthly; SPY/IWM/EFA/TLT/VNQ/GLD/DBC, T-bill filter) trailed SACEMS EW Top 2/Top 3 on gross CAGR (10.0% vs 12.6%/13.1%; Feb 2007–Sep 2026; in-sample, one path).
Using AI to Discover Calendar-based Return Anomalies
Grok-identified turn-of-the-month window (last 3 to first 3 trading days; 38% of days) averaged 0.084% daily S&P 500 return vs 0.000% outside (Dec 1927–mid-Oct 2026; AI-generated, descriptive; not a trading test).
Yield Curve as a Stock Market Indicator
10-year T-note minus 3-month T-bill spread timing (monthly; SPY to cash when spread is over 1 SD below average) underperformed buy-and-hold SPY (CAGR 9.8% ITD, 8.9% rolling 20-year vs 10.8%; SPY Jan 1993–Sep 2026; signals from Apr 1953; 0.1% switching…
Do ETFs Following Gurus/Insiders Work?
An equal-weighted, monthly-rebalanced portfolio of nine guru/insider-following ETFs underperformed SPY (10.3% vs 11.3% CAGR; both -51% MaxDD; ETF inception–Sep 2026; ignores rebalancing costs); only two of nine ETFs beat SPY's CAGR.