Everyone is excited about the potential of large language models (LLMs) to assist with forecasting, research, and countless day-to-day tasks. However, as their use expands into sensitive areas like financial prediction, serious concerns are emerging—particularly around memory leaks. In the recent paper “The Memorization Problem: Can We Trust LLMs’ Economic Forecasts?”, the authors highlight a key issue: when LLMs are tested on historical data within their training window, their high accuracy may not reflect real forecasting ability, but rather memorization of past outcomes. This undermines the reliability of backtests and creates a false sense of predictive power.

The paper “Through Stormy Seas: How Fragile is Liquidity Across Asset Classes?” is a very interesting examination of how liquidity properties have evolved over the past decade. Although the average bid–ask spread has declined, the kurtosis and skewness of the spread distribution have increased. What does this imply? On average, markets appear more liquid; however, liquidity evaporates more rapidly during stress events, amplifying tail risk and increasing execution slippage.

Hello all,

– Parameter tayloring for three Quantpedia Pro reports
– New strategy embeds
– 4th episode of our YouTube video series QuantBeats
– Reminder of the exclusive Lightspeed offer to obtain 12 FREE MONTHS of Quantpedia Premium
– 12 new Quantpedia Premium strategies have been added to our database
– 10 new related research papers have been included in existing Premium strategies during the last month
– Additionally, we have produced 7 new backtests written in QuantConnect code
– 4 new blog posts that you may find interesting have been published on our Quantpedia blog in the previous month

Corporate conferences have long been recognized as pivotal events in financial markets, serving as catalysts that signal upcoming innovations and strategic shifts. Scheduled corporate events induce market reactions that can be systematically analyzed to reveal predictable return patterns. In this work, we focus on examining the return drift exhibited by technology stocks in the days surrounding their respective conferences, employing simple quantitative methods with daily price data.

The hypothesized return drift is premised on the notion that investor sentiment and market dynamics are significantly altered by the information disseminated at these conferences. Investors, reacting to both anticipatory signals and post-announcement adjustments, tend to drive prices in a measurable manner in the windows immediately preceding, during, and after the events. By systematically analyzing stocks of companies such as Apple, Google, and Microsoft, this study aims to validate the existence of these drift patterns and shed light on the underlying mechanisms, thereby enhancing mutual understanding of event-driven asset pricing dynamics.

When using machine learning to forecast global equity returns, it’s tempting to focus on the raw prediction—whether some stock market is expected to go up or down. But our research shows that the real value lies elsewhere. What matters most isn’t the level or direction of the machine learning model’s forecast but how much it differs from a simple, price-based benchmark—such as a naive moving average signal. When that gap is wide, it often reveals hidden mispricings. In other words, it’s not about whether the ML model predicts positive or negative returns but whether its view disagrees sharply with what a basic trend-following model would suggest. Those moments of disagreement offer the most compelling opportunities for tactical country allocation.

Over the past decade and a half, the US equities have set the hard-to-beat performance benchmark. Nearly all of the other countries, no matter if small or big, emerging or developed, have lagged behind. However, what are the forces behind this outperformance? Why did most of the other markets and even investing styles bow to the US large-cap growth dominance? A new paper written by David Blitz nicely analyses the rise of the behemoth.