题目3761 · 随机过程
Choosing a Numeraire for a Benchmark-Relative Performance Fee
A fee depends on how a fund performs relative to a benchmark index. Why might the benchmark itself be a sensible numeraire candidate?
打开 →找到 23 个结果
中文题目题目4440 · 机器学习
Before Refitting More Often
What should you check first before increasing refit frequency because recent performance dipped?
打开 →题目4388 · 机器学习
Before Reporting Best Score
A time-series tuning run says a very short lookback window wins, but recent live performance has deteriorated sharply. What should you inspect first before widening the search?
打开 →题目4379 · 机器学习
Budget-Limited Tuning
Data are scarce but the search space is broad, and you want an almost unbiased performance estimate after tuning. Is nested CV conceptually appropriate here despite its cost?
打开 →题目4309 · 机器学习
Co-Adapted Hidden Units
Two hidden layers memorize pairs of co-occurring signals. In-sample metrics look great, but when one signal in the pair shifts slightly out of sample, performance collapses. Which control is most naturally aimed at reducing this co-adaptation?
打开 →题目4116 · 金融与交易
Implementation Shortfall Arithmetic 1
A buy order benchmarks performance to the arrival price 24. It fills in slices 30000@24.03, 20000@24.06, and exchange/commission fees are 0.004 dollars per share. What is the implementation shortfall in dollars and in basis points versus arrival?
打开 →题目4117 · 金融与交易
Implementation Shortfall Arithmetic 2
A sell order benchmarks performance to the arrival price 51.2. It fills in slices 25000@51.16, 15000@51.1, and exchange/commission fees are 0.003 dollars per share. What is the implementation shortfall in dollars and in basis points versus arrival?
打开 →题目4118 · 金融与交易
Implementation Shortfall Arithmetic 3
A buy order benchmarks performance to the arrival price 18.5. It fills in slices 40000@18.52, 10000@18.57, and exchange/commission fees are 0.0025 dollars per share. What is the implementation shortfall in dollars and in basis points versus arrival?
打开 →题目4119 · 金融与交易
Implementation Shortfall Arithmetic 4
A sell order benchmarks performance to the arrival price 76. It fills in slices 20000@75.95, 30000@75.9, and exchange/commission fees are 0.005 dollars per share. What is the implementation shortfall in dollars and in basis points versus arrival?
打开 →题目4120 · 金融与交易
Implementation Shortfall Arithmetic 5
A buy order benchmarks performance to the arrival price 102.4. It fills in slices 10000@102.46, 15000@102.5, 5000@102.57, and exchange/commission fees are 0.004 dollars per share. What is the implementation shortfall in dollars and in basis points versus arrival?
打开 →题目4383 · 机器学习
Noisy Metric, Narrow Budget
As model capacity increases, training performance keeps improving but validation performance stays flat. From a tuning perspective, what direction should you test next?
打开 →题目2456 · 机器学习
Row Split Instead of Issuer Split
Each issuer contributes many dated observations. Why can a random row split overstate performance even when the target is defined separately on each date?
打开 →题目1758 · 统计
Selection-on-Survivors Bias in Strategy Evaluation
A desk only records post-launch performance for strategies that first clear an internal backtest hurdle. Why does regressing realized performance on backtest score inside the launched set generally fail to recover the unconditional relationship?
打开 →题目4313 · 机器学习
When Weight Decay Starts Hurting
Performance falls as you increase weight decay. Before concluding that regularization is bad, what structural question should you ask about the signal?
打开 →题目4339 · 机器学习
Why Architecture Mismatch Hurts
Why can architecture mismatch dominate parameter count when performance is poor?
打开 →题目2593 · 机器学习
Why Averaging Cannot Cure Systematic Label Noise 20
Why can a larger forest fail to repair performance when the training labels themselves are systematically corrupted?
打开 →题目2687 · 机器学习
Why Benchmark Stability Does Not Guarantee Label Stability
Why can a model's benchmark-relative performance appear stable even while the mapping from features to returns is drifting underneath?
打开 →题目2707 · 机器学习
Why Cost Assumptions Count as Hyperparameters
Why does changing slippage curves, fee schedules, or borrow assumptions after seeing backtest performance count as extra model search?
打开 →题目2602 · 机器学习
Why Early Stopping Matters Even if Train Loss Falls 12
Why can validation performance start to deteriorate even while the training objective of boosting keeps improving?
打开 →题目2657 · 机器学习
Why Fold-to-Fold Variation Is Not a Standard Error of Deployment
Why is the standard deviation of fold scores not automatically the standard error of future production performance?
打开 →题目2720 · 机器学习
Why Live Degradation Should Be the Default Expectation
Why should a PM expect live performance to come in below the very best backtest rather than treat any shortfall as an implementation surprise?
打开 →题目2662 · 机器学习
Why Overlapping Validation Windows Complicate Score Aggregation
Why should a practitioner be careful when averaging performance over overlapping validation windows?
打开 →题目2656 · 机器学习
Why Random Row CV Breaks With Overlapping Label Horizons
Why can ordinary random row cross-validation severely overstate performance when each label depends on the next 5 trading days and adjacent rows overlap in those horizons?
打开 →