第 8 / 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
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2717Why Relaunching a Retired Strategy Can Reuse the Same LuckWhy is it dangerous to retire a strategy after disappointment and later relaunch a close cousin because a refreshed backtest looks strong again on overlapping history?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2720Why Live Degradation Should Be the Default ExpectationWhy should a PM expect live performance to come in below the very best backtest rather than treat any shortfall as an implementation surprise?机器学习困难essay未尝试面试订阅2741Bid Ceiling With a Fixed Outside OptionA trader values a data license at v. In a sealed first-price auction against one rival, winning with bid b occurs with probability p(b). If skipping the auction preserves an outside payoff of c, what inequality must b satisfy for participating with bid b to weakly dominate skipping?脑筋急转弯中等derivation未尝试面试订阅2742When a Rebate Makes Aggressive Bidding RationalIn a first-price auction, a bidder of value v receives a fixed rebate r if she loses, and nothing if she skips. If bidding b wins with probability p(b), derive the expected payoff of bidding b and compare it with skipping.脑筋急转弯中等derivation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2747All-Pay Contest Break-Even EffortTwo equally skilled quants enter an all-pay contest for a prize worth V. A contestant who spends effort e while the rival spends e 0 wins with probability e/(e+e 0), and everyone pays their own effort regardless of outcome. Holding e 0 fixed, write the payoff of effort e and solve the equation that defines break-even effort.脑筋急转弯简单derivation未尝试面试订阅2761Winner's-Curse Adjustment From a Symmetric Signal BandA common-value asset has expected resale value equal to the average of all bidders' signals. Your private signal is s, and you know that winning means your signal was the highest among a symmetric signal band around the truth. Why should your bid be below s even if E[value|signal=s]=s before conditioning on winning?脑筋急转弯简单derivation未尝试面试订阅3285Why Whitening Can Hurt Noisy Small-Eigenvalue DirectionsWhy can whitening make downstream models more fragile when some covariance eigenvalues are tiny?数学中等essay未尝试面试订阅3287Why PCA Can Miss a Predictive DirectionWhy can a low-variance direction be more useful for prediction than the first principal component?数学简单essay未尝试面试订阅3289Why Truncated SVD Is a Structured DenoiserWhy does low-rank truncation often denoise a matrix instead of merely compressing it?数学中等essay未尝试面试订阅3290Why Similar Spectra Can Hide Different SubspacesWhy do similar explained-variance ratios not guarantee that two datasets have learned the same principal directions?数学中等essay未尝试面试订阅