第 1 / 21 页
非代码面试题
显示 20 / 420 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
2565Validation Pruning With an Alpha Charge 23Replacing a single leaf by a 3-leaf subtree reduces validation loss by 4.5. If the complexity charge is alpha = 1.2 per extra leaf, should you keep the subtree?机器学习困难数值题未尝试面试订阅2671Effective Breadth Under Common Factor CorrelationA researcher tracks 100 cross-sectional alpha sleeves whose pairwise correlation is approximately 0.20. Using the rough equal-correlation breadth proxy n eff =n/(1+(n-1) ), what is the effective number of independent bets?机器学习简单数值题未尝试免费2675Break-Even Hit Rate After Trading CostsA directional model earns +1 unit on a correct trade and -1 unit on an incorrect trade before costs. Each round trip also pays a cost of 0.08 units regardless of outcome. What hit rate p makes expected net PnL zero?机器学习困难derivation未尝试面试订阅2678Net Signal-to-Cost RatioA strategy's gross expected edge is 5 bps per trade, but execution-cost uncertainty has a standard deviation of 12 bps per trade. What is the gross-edge-to-cost-noise ratio?机器学习中等数值题未尝试面试订阅2681Why Live Deployment Changes the Labeling EnvironmentWhy can a model that looked predictive in backtests become less predictive once a desk actually starts trading on it?机器学习简单essay未尝试免费2686Why Correlation Spikes Break Signal Combining RulesWhy can a portfolio of seemingly diversified alpha signals stop diversifying exactly when markets become stressed?机器学习简单essay未尝试免费2688Why Transaction Costs Are Model Uncertainty TooWhy is it wrong to treat trading costs as a fixed deduction after the predictive model has already been validated?机器学习中等essay未尝试面试订阅2690Why Fast Strategies Are More Sensitive to Small Modeling ErrorsWhy can a tiny forecast error or latency miss matter much more for a very fast strategy than for a slow rebalancing signal?机器学习困难essay未尝试面试订阅2693Why Average Error Metrics Can Hide State-Dependent FailureWhy can a model with acceptable average forecast error still be dangerous in finance?机器学习中等essay未尝试面试订阅2694Why Capacity Scaling Can Erase a Modest EdgeWhy can a signal that is clearly profitable at small size become useless when a larger book tries to deploy it?机器学习中等essay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2718Why Strategy Combination Can Also OverfitWhy does building a meta-portfolio from many individually researched strategies create another layer of overfitting risk?机器学习中等essay未尝试面试订阅2719Why Parameter Stability Matters More Than the Single Best PeakWhy is a broad plateau of good parameter values often more convincing than one spectacularly sharp optimum in a backtest heatmap?机器学习中等essay未尝试面试订阅2746Truthful Bidding Under a Capacity PenaltyIn a second-price auction for one extra inventory slot, paying above your true value v can backfire because if you win while your realized usage is low, you incur a deterministic penalty k. If the penalty is independent of the auction price and only applies when you win, what is the effective value that should replace v in the usual Vickrey truth-telling argument?脑筋急转弯困难derivation未尝试面试订阅2754Value of Tie-Breaking PriorityTwo bidders both plan to submit the same bid b in a first-price auction. If bidder A wins ties with probability q while bidder B wins ties with probability 1-q, what is the incremental expected value of A's tie-breaking priority at valuation v, relative to a fair 50-50 tie break?脑筋急转弯困难derivation未尝试面试订阅3106Half-Life When Alpha Plus Beta Equals 0.8In a GARCH-style volatility recursion, a deviation from long-run variance decays approximately by the factor = + = 4 5 each step. What is the half-life of the deviation?统计中等derivation未尝试面试订阅