EXECUTIVE PROGRAMME IN ALGORITHMIC TRADING (EPAT)
Latest Practice Exam (Updated Edition)
Question 1
A quantitative analyst fits an AR(1) model
X t =ϕ X t −1+ ϵ t
where εₜ is white noise.
Which condition guarantees covariance stationarity?
A. |φ| < 1
B. φ > 1
C. φ = 1
D. φ < −2
Answer: A
Explanation:
An AR(1) process is covariance stationary only when the autoregressive coefficient lies strictly
between −1 and +1. If |φ|≥1, shocks persist indefinitely and variance diverges.
Question 2
A strategy produces annualized returns of 18% with annualized volatility of 12%. The annual risk-free
rate is 3%.
What is the Sharpe Ratio?
A. 0.80
B. 1.00
C. 1.25
D. 1.50
Answer: C
Explanation:
Sharpe Ratio
=(18−3)/12
=15/12
=1.25
A Sharpe ratio above 1 is generally considered strong.
,EXECUTIVE PROGRAMME IN ALGORITHMIC TRADING (EPAT)
Latest Practice Exam (Updated Edition)
Question 3
Which assumption is NOT required by the classical Black-Scholes option pricing framework?
A. Constant volatility
B. Continuous trading
C. No arbitrage
D. Mean-reverting stock prices
Answer: D
Explanation:
Black-Scholes assumes geometric Brownian motion rather than mean-reverting prices. Constant
volatility, frictionless markets, and continuous hedging are core assumptions.
Question 4
A machine learning model achieves 99% training accuracy but only 61% testing accuracy.
This most likely indicates
A. Underfitting
B. High bias
C. Overfitting
D. Stationarity
Answer: C
Explanation:
The model has memorized the training data rather than learning generalizable patterns.
Question 5
Which execution algorithm attempts to match the market's historical volume profile?
A. TWAP
B. VWAP
C. Iceberg
D. Pegged Order
Answer: B
,EXECUTIVE PROGRAMME IN ALGORITHMIC TRADING (EPAT)
Latest Practice Exam (Updated Edition)
Explanation:
VWAP execution distributes trades according to expected market volume throughout the trading
session.
Question 6
Which portfolio optimization framework explicitly minimizes portfolio variance for a target expected
return?
A. Black-Litterman
B. CAPM
C. Modern Portfolio Theory
D. Kelly Criterion
Answer: C
Explanation:
Markowitz's Modern Portfolio Theory computes the efficient frontier by minimizing variance for a
given return target.
Question 7
In statistical arbitrage, cointegration primarily implies that
A. Two assets always have identical returns.
B. Individual price series are stationary.
C. A linear combination of non-stationary series is stationary.
D. Correlation equals one.
Answer: C
Explanation:
Cointegration allows mean-reversion strategies because the spread between assets remains
stationary even when individual prices follow random walks.
Question 8
What is the main benefit of vectorized backtesting compared to event-driven simulation?
A. More realistic exchange latency
B. Faster computation on historical datasets
, EXECUTIVE PROGRAMME IN ALGORITHMIC TRADING (EPAT)
Latest Practice Exam (Updated Edition)
C. Supports live trading directly
D. Eliminates look-ahead bias automatically
Answer: B
Explanation:
Vectorization leverages NumPy operations to process entire datasets efficiently, making historical
testing much faster.
Question 9
Which Greek measures an option's sensitivity to changes in implied volatility?
A. Delta
B. Gamma
C. Vega
D. Theta
Answer: C
Explanation:
Vega measures the option price change resulting from a one-percentage-point change in implied
volatility.
Question 10
The primary purpose of Principal Component Analysis in statistical arbitrage is to
A. Increase leverage
B. Remove market noise by reducing dimensionality
C. Improve execution speed
D. Price options
Answer: B
Explanation:
PCA extracts latent risk factors and removes redundant information, enabling more robust factor-
neutral trading strategies.
Question 11
When modeling volatility clustering in financial returns, which model family is most appropriate?
Latest Practice Exam (Updated Edition)
Question 1
A quantitative analyst fits an AR(1) model
X t =ϕ X t −1+ ϵ t
where εₜ is white noise.
Which condition guarantees covariance stationarity?
A. |φ| < 1
B. φ > 1
C. φ = 1
D. φ < −2
Answer: A
Explanation:
An AR(1) process is covariance stationary only when the autoregressive coefficient lies strictly
between −1 and +1. If |φ|≥1, shocks persist indefinitely and variance diverges.
Question 2
A strategy produces annualized returns of 18% with annualized volatility of 12%. The annual risk-free
rate is 3%.
What is the Sharpe Ratio?
A. 0.80
B. 1.00
C. 1.25
D. 1.50
Answer: C
Explanation:
Sharpe Ratio
=(18−3)/12
=15/12
=1.25
A Sharpe ratio above 1 is generally considered strong.
,EXECUTIVE PROGRAMME IN ALGORITHMIC TRADING (EPAT)
Latest Practice Exam (Updated Edition)
Question 3
Which assumption is NOT required by the classical Black-Scholes option pricing framework?
A. Constant volatility
B. Continuous trading
C. No arbitrage
D. Mean-reverting stock prices
Answer: D
Explanation:
Black-Scholes assumes geometric Brownian motion rather than mean-reverting prices. Constant
volatility, frictionless markets, and continuous hedging are core assumptions.
Question 4
A machine learning model achieves 99% training accuracy but only 61% testing accuracy.
This most likely indicates
A. Underfitting
B. High bias
C. Overfitting
D. Stationarity
Answer: C
Explanation:
The model has memorized the training data rather than learning generalizable patterns.
Question 5
Which execution algorithm attempts to match the market's historical volume profile?
A. TWAP
B. VWAP
C. Iceberg
D. Pegged Order
Answer: B
,EXECUTIVE PROGRAMME IN ALGORITHMIC TRADING (EPAT)
Latest Practice Exam (Updated Edition)
Explanation:
VWAP execution distributes trades according to expected market volume throughout the trading
session.
Question 6
Which portfolio optimization framework explicitly minimizes portfolio variance for a target expected
return?
A. Black-Litterman
B. CAPM
C. Modern Portfolio Theory
D. Kelly Criterion
Answer: C
Explanation:
Markowitz's Modern Portfolio Theory computes the efficient frontier by minimizing variance for a
given return target.
Question 7
In statistical arbitrage, cointegration primarily implies that
A. Two assets always have identical returns.
B. Individual price series are stationary.
C. A linear combination of non-stationary series is stationary.
D. Correlation equals one.
Answer: C
Explanation:
Cointegration allows mean-reversion strategies because the spread between assets remains
stationary even when individual prices follow random walks.
Question 8
What is the main benefit of vectorized backtesting compared to event-driven simulation?
A. More realistic exchange latency
B. Faster computation on historical datasets
, EXECUTIVE PROGRAMME IN ALGORITHMIC TRADING (EPAT)
Latest Practice Exam (Updated Edition)
C. Supports live trading directly
D. Eliminates look-ahead bias automatically
Answer: B
Explanation:
Vectorization leverages NumPy operations to process entire datasets efficiently, making historical
testing much faster.
Question 9
Which Greek measures an option's sensitivity to changes in implied volatility?
A. Delta
B. Gamma
C. Vega
D. Theta
Answer: C
Explanation:
Vega measures the option price change resulting from a one-percentage-point change in implied
volatility.
Question 10
The primary purpose of Principal Component Analysis in statistical arbitrage is to
A. Increase leverage
B. Remove market noise by reducing dimensionality
C. Improve execution speed
D. Price options
Answer: B
Explanation:
PCA extracts latent risk factors and removes redundant information, enabling more robust factor-
neutral trading strategies.
Question 11
When modeling volatility clustering in financial returns, which model family is most appropriate?
