Analysis
What does out-of-sample mean in a forecasting paper?
Out-of-sample scoring is the only published test that approximates live trading, and most papers still manage to overstate it.
Author
Sofia Lindqvist
Sofia Lindqvist builds models for a living and is unusually honest about how often they are wrong. She covers portfolio construction and rebalancing, model validation design, payment fraud methods and where liability lands, media formats assessed by cost and lifespan, and wages, hours and bargaining across employment sectors.
Articles by Sofia Lindqvist
Analysis
What replaced the guidance that has governed bank models since 2011?
Federal regulators rewrote model risk management guidance in April 2026, keeping the same validation core built around conceptual soundness and outcomes…
Analysis
The Fed's stress tests changed shape: what averaging does to comparability
The 2026 results — 32 banks, $708 billion in projected losses — arrive inside a methodology overhaul that trades year-to-year volatility for smoother…
Analysis
How to evaluate an alternative data vendor: a working checklist
Point-in-time integrity, provenance, history depth and decay tests — the questions that separate a signal from a spreadsheet.
Analysis
What regime-detection models can and cannot do
Hidden Markov and clustering models summarize states that already happened — timely labeling of live regimes is the hard, partly unsolved part.