第 18 / 37 页
非代码面试题
显示 20 / 738 道匹配题目
答题状态：未尝试未正确已正确
ID题目领域难度题型进度权限
4403Rolling Mean That PeeksA feature at time t uses a rolling mean computed from t-19 through t+1. Why is that unacceptable even if it is only one extra day?机器学习中等essay未尝试面试订阅4404Cross-Sectional Residualization TimingWhy should factor residualization for a daily return feature use exposures known at the feature timestamp rather than exposures estimated using later returns?机器学习中等essay未尝试面试订阅4405Corporate Action MisalignmentA feature uses raw prices while the target uses split-adjusted future returns. What problem can that create?机器学习中等essay未尝试面试订阅4406Longer Horizon, More SmoothingIf you lengthen a forward-return label from 1 day to 20 days while sampling daily, what happens to overlap and effective sample independence?机器学习中等essay未尝试面试订阅4407Residual Feature DriftWhy can a residualized return feature that looked stable in-sample become unstable after a regime shift?机器学习中等essay未尝试面试订阅4408Normalization Window LengthIf you shorten the rolling window used to z-score a return feature, how does the feature usually respond to recent shocks?机器学习中等essay未尝试面试订阅4409Raw Returns Versus Scaled ReturnsWhy can a model trained on raw multi-asset returns misallocate attention across assets compared with standardized return features?机器学习中等essay未尝试面试订阅4410Sampling FrequencyIf you downsample from daily to weekly observations when using medium-horizon return features, what usually happens to overlap and microstructure noise?机器学习中等essay未尝试面试订阅4411Before Adding A FeatureBefore adding a new return-based feature to your model, what is the first alignment question you should ask?机器学习中等essay未尝试面试订阅4412Before ResidualizingWhat should you clarify before residualizing returns against factors and treating the residual as a new feature?机器学习中等essay未尝试面试订阅4413Before Expanding HorizonWhat should you check first before lengthening a forward-return horizon because the 1-day target looks noisy?机器学习中等essay未尝试面试订阅4414Before StandardizingBefore standardizing a return feature, what should you check about the universe you are mixing?机器学习中等essay未尝试面试订阅4415Before Trusting Feature ImportanceA return feature looks very important in a trained model. What should you check first before concluding it captures genuine alpha?机器学习中等essay未尝试面试订阅4426Rolling Versus ExpandingA market regime has shifted several times in the last three years. Why can a rolling walk-forward window be more informative than a purely expanding window?机器学习中等essay未尝试面试订阅4427Refit CadenceWhy can refitting every single day be worse than refitting monthly, even when more frequent updating sounds adaptive?机器学习中等essay未尝试面试订阅4428Embargo IntuitionWhy is a time embargo helpful when labels depend on future returns that overlap across neighboring samples?机器学习中等essay未尝试面试订阅4429Walk-Forward Is Not MagicWhy does a clean walk-forward protocol still not guarantee that a strategy will survive live trading?机器学习中等essay未尝试面试订阅4430Comparing Window SchemesTwo walk-forward schemes give different validation results. What is the first structural question you should ask before deciding one is 'better'?机器学习中等essay未尝试面试订阅4431Longer Train WindowIf you lengthen the training window in walk-forward validation, what tradeoff usually changes?机器学习中等essay未尝试面试订阅4432Longer Test WindowIf you lengthen the test block while holding the train block fixed, what usually happens to score variance and regime purity?机器学习中等essay未尝试面试订阅