July 2026

Model selection on multiply imputed data is biased toward the candidates with more missing information, which in the nested families studied are the more complex models, because their larger relative increase in variance makes the fit look better than it is. We trace this to a deviance bias in the...

We study high-probability tracking of a Behavioural strategy that drifts over time in an imperfect-information extensive-form game, when the strategy is observed only through a known, possibly noisy emission channel at each recognized visit to an information set.

Football is widely acknowledged as a global cultural phenomenon that transcends athletic competition, shaping identities, public opinion, and social cohesion. This exploratory study investigates faculty perceptions of the England–Argentina match through informal digital surveys conducted via WhatsApp. Fifteen faculty members from a School of Languages and Business in Guayaquil, Ecuador,...

Large language models now write code faster than people can meaningfully read it, yet the dominant acceptance interface for AI-generated code is still line-by-line human review. We argue that this interface neither scales nor reliably catches plausible-but-wrong programs, and that the artifact a human signs off on should change: from...