Textbook Quiz – Week 3

Answer each question in your own words. Be specific.

Your full name *0 / 150

Your institutional email address *Must match the address on your course enrolment. Checked against the roster.

Your class *Used to keep each class cohort’s records separate.

Which review period is this? *Submit once per period. Earlier submissions are kept, not replaced.

1. A platform offers several million possible factor definitions created from different fields, transformations, universes, horizons, and weighting schemes. A researcher selects the factor with the largest in-sample t-statistic. Which research design provides the strongest protection against overfitting? A) Average the 100 largest in-sample t-statistics without further testing. B) Accept the largest t-statistic because a very large candidate set ensures that the best factor has been found. C) Continue adjusting the factor until it performs well on the final test sample. D) Constrain the search, control multiple testing, and reserve untouched validation Type the letter (A, B, C, or D) of the best answer. *0 / 10

2. A portfolio must buy 600,000 shares during three trading intervals. Expected market volume is 1,000,000, 2,000,000, and 1,000,000 shares. The trader wants a constant 15% participation rate in each interval. What schedule should be used? A) 200,000, 200,000, 200,000 B) 0, 600,000, 0 C) 300,000, 150,000, 150,000 D) 150,000, 300,000, 150,000 Type the letter (A, B, C, or D) of the best answer. *0 / 10

3. A university endowment needs a 15-year policy portfolio but also wants protection against a possible six-month inflation shock. Which architecture is most appropriate? A) Use strategic allocation for the long horizon and a governed tactical overlay B) Use the six-month inflation model to determine the entire 15-year allocation. C) Ignore the inflation threat because tactical forecasting is always less accurate than strategic forecasting. D) Replace all forecasts with historical average returns from the most recent six months. Type the letter (A, B, C, or D) of the best answer. *0 / 10

4. A consultant estimates long-run equity returns using the current index constituents' histories over the past 25 years. Companies that failed, merged, or were deleted are absent. Which correction is most important? A) Increase the historical mean to compensate for missing observations. B) Rebuild point-in-time membership with delistings and failures C) Use current constituents but equal-weight them. D) Replace all individual stocks with the current capitalization-weighted index. Type the letter (A, B, C, or D) of the best answer. *0 / 10

5. A signal is calculated immediately after the market close on day t. Orders can first be executed at the opening price on day t+1. The desired response is the return earned through the close of day t+5. Which target is properly aligned? A) P(t+5, close) / P(t+1, open) – 1 B) P(t+5, close) / P(t, close) – 1 C) P(t+1, open) / P(t, close) – 1 D) P(t+5, open) / P(t+5, close) – 1 Type the letter (A, B, C, or D) of the best answer. *0 / 10

6. For asset i on date t, the feature vector is X(i,t). The intended forecast is the asset's return over the next period, r(i,t+1). Which row construction is valid? A) Pair X(i,t) with r(i,t). B) Pair the cross-sectional average of X(t) with each asset's contemporaneous return. C) Pair X(i,t+1) with r(i,t+1). D) Pair X(i,t) with r(i,t+1) and keep chronological train-test separation Type the letter (A, B, C, or D) of the best answer. *0 / 10

7. Under the performance-fee reset incentive model, a manager is paid a performance fee, is currently below the hurdle, and is close to the date at which the performance account resets. Which behavioral outcome does the model associate with this state? A) Excess risk: incentive to gamble before reset B) Immediate liquidation regardless of the portfolio's expected return C) Appropriate risk because the hurdle is irrelevant D) Too little risk because the manager locks in the current result Type the letter (A, B, C, or D) of the best answer. *0 / 10

8. A model attempts to predict a manager's next risk decision using only the most recent portfolio return. Compensation, however, depends on a high-water mark, a hurdle, and the time remaining until bonus or reset. Which state representation is most appropriate? A) NAV, distance to hurdle/HWM, fee terms, and time to reset B) The manager's historical average return alone C) Current market capitalization and trading volume D) Current return alone Type the letter (A, B, C, or D) of the best answer. *0 / 10

9. An equity classifier has strong out-of-sample ranking ability: high-score stocks outperform low-score stocks. The score is not calibrated as an expected return, and no evidence shows that a score of 0.8 warrants twice the position of a score of 0.4. What is the best conclusion? A) Convert the scores into volatilities and optimize without further testing. B) Use score-proportional weights because ranking accuracy proves sizing accuracy. C) Square the scores to emphasize the strongest predictions. D) Treat it as a ranking model; build a separate sizing rule Type the letter (A, B, C, or D) of the best answer. *0 / 10

10. A model reliably identifies which securities should be long and which should be short. It does not reliably forecast return magnitude or optimal holding period. The portfolio must rebalance weekly. Which implementation is best supported? A) Use the model score to determine leverage, timing, and stop-loss thresholds simultaneously. B) Size positions in direct proportion to the raw model score and allow holding periods to vary without limit. C) Equal-weight or risk-normalize validated selections on schedule D) Do not trade because a useful model must solve selection, direction, timing, and quantity jointly. Type the letter (A, B, C, or D) of the best answer. *0 / 10

Confirmation *I confirm that this submission is accurate and may be used for course administration.