Investing Expertise

How is the growing use of large language models (LLM) in production of academic papers in finance affecting outputs? In their June 2025 paper entitled “Certainly! Generative AI and its Impact on Academic Writing (in Finance)”, Thomas Walther and Marie Dutordoir investigate how use of LLMs has affected academic writing in finance, with focus on readability. They quantify readability via the widely-used Flesch- Kincaid Index (FKI) and Gunning-Fog Index (GFI), both of which estimate the number of years of education required to understand a certain text in a first reading. They also look at author characteristics associated with LLM use and the impact of LLM use on author publication quantity, quality (journal rankings) and impact (citations). They treat the release of ChatGPT at the end of November 2022 as the LLM adoption date. To determine author use of LLMs, they: (1) identify specific words that disproportionally appear in LLM-generated text (AI Words); (2) count the AI Words in the abstract, introduction and full text of each article; and, (3) divide the number of AI Words by the total number of words in the article. Using 41,489 articles (minimum 200 words) from 34 finance journals published in English from January 1, 2000 to April 1, 2025 (32,993 from before December 1, 2022 and 8,496 after), they find that: Keep Reading

Do the trading activities of especially informed equity and equity option traders predict stock returns? In the June 2025 revision of their paper entitled “An Information Factor: What Are Skilled Investors Buying and Selling?”, Matthew Ma, Xiumin Martin, Matthew Ringgenberg and Guofu Zhou construct an information factor (INFO) using the trades of corporate insiders, short sellers and option traders. Specifically, they each month for each stock calculate:

• To inform the long side of the INFO factor portfolio, net insider purchases (purchases minus sales).
• To inform the short side of the INFO factor portfolio:
  • Short interest (number of shares shorted divided by shares outstanding).
  • Option trading (total option volume divided by total stock volume).
• For each of these three metrics, assign a rank from 1 to 100, with higher rank indicating higher level of positive private information.
• Average the three ranks to compute an information score.
• Reform 10 equal-weighted (decile) portfolios of stocks sorted by information score, with the INFO factor portfolio long the top decile and short the bottom.
• Hold the portfolios for one month.

They assess the impact of stock trading frictions by assuming costs equal to half the respective effective bid-ask spreads. Using insider trading, short interest and option/stock trading volumes during January 1996 through December 2019, they find that: Keep Reading

Could mutual fund managers achieve performance improvements by “hiring” artificial intelligence (AI) analysts? In their May 2025 working draft entitled “The Shadow Price of ‘Public’ Information”,  Ed deHaan, Chanseok Lee, Miao Liu and Suzie Noh estimate the value of an AI stock picking analyst by having it exploit public data to improve mutual fund holdings. Specifically, they:

• Each year train the AI (random forest model) on an expanding window of 170 stock market, accounting, analysts and macroeconomic inputs.
• For each mutual fund each quarter:
  • Have the AI make limited improvement/replacement picks similar to the type of stocks already in the fund, with the goal of maximizing benchmark-adjusted returns. The benchmark for each fund consists of a portfolio of stocks with similar characteristics.
  • Require the AI to mimic original fund and holdings sizes.
  • Quantify the profit difference between original and AI-improved portfolios.

They also consider matched portfolios selected completely (not partial replacement) by the AI analyst. Using data for 3,337 active/diversified U.S. equity mutual funds during 1990 through 2020, and for firm characteristics and trading data of all U.S. listed common stocks, macroeconomic variables, analyst forecasts, credit ratings and market/firm sentiment data starting 1980, they find that:

Keep Reading

Do funds based on holdings/trades of members of the U.S. Congress and their families beat the market? To investigate, we look at performances of two recently introduced exchange-traded funds (ETF):

1. Unusual Whales Subversive Democratic Trading ETF (NANC) – invests primarily in stocks held by Democratic members of Congress and/or their families per public disclosure filings.
2. Unusual Whales Subversive Republican Trading ETF (KRUZ) – invests primarily in stocks held by Republican members of Congress and/or their families per public disclosure filings.

We use SPDR S&P 500 ETF Trust (SPY) as a benchmark. Using monthly dividend-adjusted prices for NANC, KRUZ and SPY during February 2023 (NANC and KRUZ inception) through May 2025, we find that: Keep Reading

How well do the short-term stock picks of publicly available artificial intelligence (AI) platforms perform? To investigate, we asked Grok, ChatGPT, Perplexity, Gemini and Meta AI the following questions on April 20, 2025:

• Please succinctly provide your unique best long idea for the next 30 days.
• Please succinctly provide your unique best shorting idea for the next 30 days.

We then: (1) calculated total returns for the resulting stock picks from the close on April 21, 2025 to the close on May 21, 2025; (2) averaged the returns for long and short picks; and, (3) compared  average returns for long and short picks. We include total returns for SPDR S&P 500 ETF (SPY) and Invesco QQQ Trust (QQQ) over this same interval for context. Using dividend-adjusted prices for the specified picks, we find that:

Keep Reading

Should investors, particularly those employing machine learning, prefer complex or simple prediction models? In the May 2025 revision of his paper entitled “Simplified: A Closer Look at the Virtue of Complexity in Return Prediction”, Daniel Buncic challenges prior research finding that portfolio performance (Sharpe ratio) increases with machine learning model complexity when the number of inputs (potential predictors) greatly exceeds the number of training observations. Using the same dataset, prediction models and portfolio evaluation methods as the prior research, he finds that: Keep Reading

We occasionally ask publicly available artificial intelligence (AI) platforms for investing ideas and post results on the CXOAdvisory X account. Two recent examples are:

1. “Please concisely provide your unique choice for the best risk-adjusted investment for generating monthly or quarterly income.”
2. “Using data up to now, please concisely provide your unique estimate of which asset class will be strongest and which will be weakest over the next three months.”

We may use responses of such items to assess usefulness of AIs as investing aids, such as in a CXOAdvisory.com post scheduled for later this week.

Can large language models (LLM) be trusted for economic/financial forecasts during periods within their training data? In their April 2025 paper entitled “The Memorization Problem: Can We Trust LLMs’ Economic Forecasts?”, Alejandro Lopez-Lira, Yuehua Tang and Mingyin Zhu evaluate use of  ChatGPT 4o (knowledge cutoff October 2023) for economic/financial forecasting via:

• Forecasts of variables before and after knowledge cutoff.
• Explicit instructions to ignore knowledge during periods before the cutoff.
• Masking of inputs (anonymized firm names or dates) to mitigate use of memorized data in forecasts before knowledge cutoff.

Using data for major economic indicators, stock index levels, individual stock returns/conference calls and Wall Steet Journal (WSJ) headlines during December 1989 through February 2025, they find that: Keep Reading

Should stock return model complexity guide breadth of input data? In their May 2025 paper entitled “Model Complexity and the Performance of Global Versus Regional Models”, Minghui Chen, Matthias Hanauer and Tobias Kalsbach assess the predictive performance of global versus regional inputs for stock return models based on linear and machine learnings algorithms: ordinary least squares regression (OLS); elastic net (ENET); random forest (RF); gradient-boosted regression trees (GBRT); and, neural networks (NN). Monthly model inputs include 36 firm-level characteristics and associated stock trading data in U.S. dollars for 24 developed market countries, suppressing effects of megacaps and excluding microcaps (the smallest stocks per country comprising 3% of overall market capitalization).  They segment country markets into four regions: North America, Europe, Japan and Asia Pacific. Model training employs an expanding window (initially six years, extended year by year), followed by a 6-year validation interval and a 1-year test interval. For each model, each month, they reform a portfolio that is long (short) the fifth, or quintile, of stocks with the highest (lowest) predicted returns. Using the specified monthly firm/stock inputs during July 1990 through December 2021, they find that:

Keep Reading

Can investors exploit analyst stock price targets by finding the best analysts and overweighting the most extreme target-implied returns? In their March 2025 paper entitled “Alpha in Analysts”, Álvaro Cartea and Qi Jin test the informativeness and exploitability of sell-side analyst stock price targets. To test informativeness of target prices, they each month for each analyst:

• Use price targets to deduce 12-month return forecasts.
• Form a hedge portfolio that is long (short) stocks with positive (negative) return forecasts, with weights proportional to magnitudes of forecasted returns and absolute value of the sum of weights equal to one.
• Compare analyst portfolio performance to that of an equal-weighted, long-only portfolio of the same stocks.

To test exploitability of results, they each month:

• Predict portfolio profitability for each analyst via an inception-to-date regression of six analyst performance metrics up to 12 months ago (capturing historical performance and breadth of stock coverage) versus next-month portfolio return.
• Construct a portfolio of analyst portfolios with higher (lower) allocations to those with higher (lower) predicted returns.

Using daily analyst price targets and associated stock returns/firm characteristics as available for common NYSE/AMEX/NASDAQ stocks during January 1999 through November 2024, they find that: Keep Reading