第 1 / 2 页
非代码面试题
显示 20 / 24 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
2696False Strategy Surviving Two Independent Research GatesA desk tries 80 genuinely null strategy ideas. A strategy is kept only if it passes an in-sample screen at 10% and then a fresh out-of-sample confirmation at 5%, with the two tests treated as independent under the null. What is the probability at least one null idea survives both gates?机器学习简单derivation未尝试面试订阅2697False Winner Probability After Clustering 240 Variants into 24 FamiliesA researcher generates 240 heavily correlated strategy variants but argues they amount to only 24 effectively independent families. If the desk still flags any family with p-value below 8%, what is the approximate probability of at least one false family-level winner under the null?机器学习简单数值题未尝试面试订阅2698Single-Test Significance Needed to Cap Familywise False Discovery at 10%A desk plans to try 60 effectively independent strategy ideas. What single-test significance level alpha keeps the probability of at least one false winner at exactly 10% under the independence approximation?机器学习中等数值题未尝试面试订阅2699Null Survival Through Two Independent ScreensA null strategy must first pass an exploratory screen at level 10% and then pass a final untouched holdout at level 5%. If the two tests are independent under the null, what is the probability that a null strategy survives both?机器学习中等derivation未尝试面试订阅2700Familywise Error Using Effective Clusters Instead of Raw VariantsA researcher produced 120 parameter variants, but they cluster into roughly 6 effectively independent families. Using a 5% per-family threshold as a rough approximation, what is the probability of at least one false winner across the 6 families?机器学习困难数值题未尝试面试订阅2701Search Depth Hidden in Design KnobsA team tried 4 universes, 5 rebalance frequencies, and 6 transaction-cost assumptions before reporting the best result. How many distinct design combinations were implicitly searched?机器学习简单数值题未尝试面试订阅2702Chance the Best Null t-Statistic Exceeds 2.4 Across 50 VariantsSuppose 50 genuinely null standardized t-statistics are approximately independent N(0,1). What is the probability the largest of them exceeds 2.4?机器学习中等数值题未尝试面试订阅2703Probability That a Null Sharpe Sweep Produces a Big WinnerSuppose 25 independent null strategies each produce an in-sample Sharpe that is approximately standard normal. What is the probability that the best observed Sharpe exceeds 1.5?机器学习中等derivation未尝试面试订阅2704Expected Null Strategies Surviving a Screening FunnelA research platform runs 200 null strategies. Only strategies with in-sample p-value below 15% are promoted, and each promoted strategy must then pass a fresh 5% confirmation test. Assuming independence under the null, what is the expected number of false strategies that survive both stages?机器学习中等数值题未尝试面试订阅2706Why an Untouched Holdout Stops Being UntouchedWhy does a final holdout lose its evidential value once researchers repeatedly inspect it during idea iteration?机器学习简单essay未尝试面试订阅2707Why Cost Assumptions Count as HyperparametersWhy does changing slippage curves, fee schedules, or borrow assumptions after seeing backtest performance count as extra model search?机器学习简单essay未尝试面试订阅2708Why Universe Choice Is Part of the Search TreeWhy should changing the tradable universe be counted as another research branch rather than as harmless context?机器学习中等essay未尝试面试订阅2709Why Ranking by In-Sample Sharpe Prefers Noise PeaksWhy does selecting the strategy with the highest in-sample Sharpe systematically bias the chosen strategy upward even when all candidates are mediocre?机器学习困难essay未尝试面试订阅2710Why CPCV Helps but Does Not Solve Adaptive Idea GenerationWhy can combinatorial pathwise validation improve robustness checks without fully solving the problem of researchers inventing new ideas after seeing the old results?机器学习困难essay未尝试面试订阅2711Why Paper Trading Can Be More Informative Than One More BacktestWhy can a period of forward paper trading add more evidence than squeezing one more clever slice out of the historical sample?机器学习简单essay未尝试面试订阅2712Why Many Small Tweaks Still Count as Deep SearchWhy is it misleading to claim that no serious overfitting occurred just because the final strategy differs from the baseline by only many small tweaks?机器学习中等essay未尝试面试订阅2713Why High-Turnover Strategies Are Easier to OverfitWhy does backtest overfit become especially dangerous for very high-turnover strategies?机器学习中等essay未尝试面试订阅2714Why Search Depth Is Bigger Than the Number of Named StrategiesWhy can a team that claims to have tested only five named strategies still have conducted a much deeper search than that number suggests?机器学习困难essay未尝试面试订阅2715Why Economic Logic Acts Like a Prior Against NoiseWhy does a strategy with a credible economic mechanism deserve more trust than a statistically similar strategy with no coherent story?机器学习困难essay未尝试面试订阅2716Why Stop-Loss Tuning Is Also Multiple TestingWhy does picking a stop-loss threshold after looking at the full historical equity curve count as backtest search rather than risk management hygiene?机器学习简单essay未尝试免费