Tag
#research
9 articles
-
Common Backtesting Mistakes That Make Results Look Better Than Reality
The most common backtesting mistakes are future information leaking into past decisions, survivorship in the universe, ignored costs, and testing so many variations that something looks good by luck.
Read article -
In-Sample vs Out-of-Sample Testing: The Minimum Honest Backtest Check
In-sample data is where a strategy is built and tuned. Out-of-sample data is held back and used once to judge it. Splitting the two is the least you can do to avoid fooling yourself.
Read article -
What Is Overfitting in Backtesting? Curve-Fitting, Parameters and How to Detect It
Overfitting in backtesting is tuning a strategy until it describes the noise in one sample of history rather than any durable pattern. It looks like a great result and behaves like a coin flip.
Read article -
From Reading Documents to Asking Questions
Research is shifting from reading whole filings front to back to interrogating them with specific questions and getting sourced answers, which changes where an analyst spends time and attention.
Read article -
How AI Compresses a Week of Research Into an Hour
AI collapses the grunt work of primary research, gathering, reading, and spreading numbers, from days to minutes. The judgement, the part that decides the outcome, still takes a human.
Read article -
Information Overload Is the Real Edge Killer
More information is not better research. The edge is synthesis and focus: knowing the few variables that matter for each holding, writing them down, and ignoring the rest.
Read article -
The Death of Ctrl+F in Annual Reports
Keyword search finds strings, not meaning. It misses synonyms, ignores context, and cannot answer a question, which is why reading filings is shifting from searching words to asking questions.
Read article -
Why Every Analyst Will Have an AI Associate
An AI associate does the tireless first pass, pulling numbers and reading every page, while the human analyst keeps the judgement, conviction, and accountability.
Read article -
Why Point-in-Time Data Matters in Research and Backtests
Point-in-time data means using the numbers that were actually knowable on a given date, not today's restated version. Skip it and your research quietly looks smarter than it was.
Read article