题目2681 · 机器学习
Why Live Deployment Changes the Labeling Environment
Why can a model that looked predictive in backtests become less predictive once a desk actually starts trading on it?
打开 →找到 5 个结果
中文题目题目2756 · 脑筋急转弯
Bidder Utility Under Revenue Equivalence
A bidder has value $0.9$ in a two-bidder private-value environment with values i.i.d. $\mathrm{Uniform}(0,1)$. Compute that bidder's expected utility in (a) the first-price equilibrium and (b) the truthful second-price auction. Verify that the utilities agree.
打开 →题目1723 · 统计
Underpowered Winner’s Curse
A sparse signal library was screened on short samples; the surviving signal is significant but came from a very low-power environment. Why should its in-sample effect size be viewed skeptically?
打开 →题目3229 · 统计
When Large-Sample Frequentist Asymptotics Are Attractive
Give one practical reason a quant team might prefer a frequentist asymptotic analysis over a full Bayesian analysis when the sample is enormous and arrives from a stable high-liquidity environment.
打开 →题目2765 · 脑筋急转弯
Why Private Values Avoid the Winner's Curse
Explain why the winner's curse is a feature of common-value auctions but not of independent private-value auctions. Your answer should make clear what information winning reveals in each environment.
打开 →