ISYE 6402 MIDTERM EXAM FULL LENGTH TEST BANK 2025/2026 |
ACCURATE CURRENTLY TESTING EXAM VERSIONS WITH ACTUAL
QUESTIONS AND ANSWERS WITH A RATIONALES AND A STUDY
GUIDE| LATEST UPDATE
Question 1
What is the defining characteristic of a time series dataset?
A) It is a random sample of independent observations.
B) It is a sequence of data points indexed in time order.
C) It is data collected from different subjects at the same point in time (cross-sectional).
D) It must contain at least two variables to show correlation.
E) It always follows a normal distribution.
Correct Answer: B) It is a sequence of data points indexed in time order.
Rationale: The fundamental feature of a time series is its temporal ordering. The sequence
of the data points is critical, as the primary goal is often to analyze the dependencies and
patterns that evolve over time. Unlike a random sample, the order of observations is not
interchangeable.
Question 2
A time series is said to be weakly stationary (or covariance stationary) if which of the following
properties are constant over time?
A) The mean and the variance only.
B) The entire joint probability distribution of the observations.
C) The mean, variance, and autocovariance for each fixed lag.
D) The autocorrelations (ACF) and partial autocorrelations (PACF).
E) The mean and the partial autocorrelations (PACF) only.
Correct Answer: C) The mean, variance, and autocovariance for each fixed lag.
Rationale: Weak stationarity is defined by three conditions: 1) The mean of the series is
constant and does not depend on time (E[Xₜ] = μ). 2) The variance of the series is constant
and does not depend on time (Var(Xₜ) = σ²). 3) The autocovariance between any two
observations depends only on the lag (the time difference) between them, not on the specific
time t (Cov(Xₜ, Xₜ₊ₕ) = γₕ).
,[Type here]
Question 3
Which of the following is a characteristic of a white noise process?
A) The observations are correlated with a constant mean and variance.
B) The observations have a constant mean of zero, a constant variance, and are serially
uncorrelated.
C) The observations have a mean of zero but a changing variance.
D) The observations are serially correlated with a zero mean.
E) The observations follow a normal distribution with a non-zero mean.
Correct Answer: B) The observations have a constant mean of zero, a constant variance,
and are serially uncorrelated.
Rationale: A white noise process is the building block of many time series models. It is a
sequence of random variables with a mean of zero, a constant variance, and zero
autocovariance for all non-zero lags. This means that each observation is independent of all
others.
Question 4
What does the Autocorrelation Function (ACF) measure?
A) The correlation between a time series and a different time series.
B) The correlation between a time series and its own past values at different lags.
C) The correlation between a time series and its future values.
D) The partial correlation between observations after removing the effect of intervening lags.
E) The variance of the time series at different points in time.
Correct Answer: B) The correlation between a time series and its own past values at
different lags.
Rationale: The ACF at lag k, denoted ρ(k), measures the linear relationship between an
observation at time t (Xₜ) and the observation k periods in the past (Xₜ₋ₖ). It is a key tool
for identifying the structure of a time series.
Question 5
What does the Partial Autocorrelation Function (PACF) at lag k measure?
A) The correlation between Xₜ and Xₜ₋ₖ after removing the linear dependence on the
intermediate observations (Xₜ₋₁, Xₜ₋₂, ..., Xₜ₋ₖ₊₁).
,[Type here]
B) The total correlation between Xₜ and Xₜ₋ₖ.
C) The correlation between the first k and the last k observations of a series.
D) A moving average of the autocorrelations up to lag k.
E) The probability that the autocorrelation at lag k is significant.
Correct Answer: A) The correlation between Xₜ and Xₜ₋ₖ after removing the linear
dependence on the intermediate observations (Xₜ₋₁, Xₜ₋₂, ..., Xₜ₋ₖ₊₁).
Rationale: The PACF provides a "cleaner" view of the direct relationship between two
observations by removing the confounding effects of the observations that lie between
them. It is crucial for identifying the order of autoregressive (AR) models.
Question 6
Classical decomposition of a time series typically breaks it down into which four components?
A) Mean, Variance, Skewness, and Kurtosis
B) Trend, Seasonality, Cyclical, and Irregular (Noise)
C) Autoregressive, Moving Average, Integrated, and Seasonal
D) Level, Slope, Period, and Error
E) Past, Present, Future, and Random
Correct Answer: B) Trend, Seasonality, Cyclical, and Irregular (Noise)
Rationale: The classical decomposition model assumes that a time series can be broken
down into these four underlying components: the long-term Trend, the fixed-period
Seasonality, the non-fixed-period Cyclical fluctuations, and the random, unpredictable
Irregular component.
Question 7
What is the primary purpose of using a moving average to smooth a time series?
A) To make the series stationary.
B) To forecast future values with perfect accuracy.
C) To highlight the underlying trend and remove short-term fluctuations or noise.
D) To identify the order of an ARMA model.
E) To calculate the partial autocorrelations of the series.
, [Type here]
Correct Answer: C) To highlight the underlying trend and remove short-term fluctuations
or noise.
Rationale: A moving average works by replacing each data point with the average of itself
and a certain number of its neighbors. This averaging process smooths out the random,
short-term "jumps" in the data, making the longer-term trend or pattern more visible.
Question 8
A time series analyst observes an Autocorrelation Function (ACF) plot that shows a gradual,
exponential decay, and a Partial Autocorrelation Function (PACF) plot that cuts off abruptly after
lag 1. What is the most likely model for this time series?
A) An MA(1) model
B) An AR(1) model
C) An ARMA(1,1) model
D) A white noise process
E) An MA(2) model
Correct Answer: B) An AR(1) model
Rationale: This is the classic "signature" of an Autoregressive model of order 1 (AR(1)). An
AR(p) process is characterized by a PACF that cuts off after lag p and an ACF that tails off
(decays exponentially or with a sinusoidal pattern).
Question 9
A time series analyst observes an Autocorrelation Function (ACF) plot that cuts off abruptly after
lag 2. The Partial Autocorrelation Function (PACF) plot appears to tail off exponentially. What is
the most likely model for this time series?
A) An AR(2) model
B) An ARMA(2,2) model
C) An MA(2) model
D) An ARIMA(2,1,0) model
E) A white noise process
Correct Answer: C) An MA(2) model
Rationale: This is the classic "signature" of a Moving Average model of order 2 (MA(2)).
ACCURATE CURRENTLY TESTING EXAM VERSIONS WITH ACTUAL
QUESTIONS AND ANSWERS WITH A RATIONALES AND A STUDY
GUIDE| LATEST UPDATE
Question 1
What is the defining characteristic of a time series dataset?
A) It is a random sample of independent observations.
B) It is a sequence of data points indexed in time order.
C) It is data collected from different subjects at the same point in time (cross-sectional).
D) It must contain at least two variables to show correlation.
E) It always follows a normal distribution.
Correct Answer: B) It is a sequence of data points indexed in time order.
Rationale: The fundamental feature of a time series is its temporal ordering. The sequence
of the data points is critical, as the primary goal is often to analyze the dependencies and
patterns that evolve over time. Unlike a random sample, the order of observations is not
interchangeable.
Question 2
A time series is said to be weakly stationary (or covariance stationary) if which of the following
properties are constant over time?
A) The mean and the variance only.
B) The entire joint probability distribution of the observations.
C) The mean, variance, and autocovariance for each fixed lag.
D) The autocorrelations (ACF) and partial autocorrelations (PACF).
E) The mean and the partial autocorrelations (PACF) only.
Correct Answer: C) The mean, variance, and autocovariance for each fixed lag.
Rationale: Weak stationarity is defined by three conditions: 1) The mean of the series is
constant and does not depend on time (E[Xₜ] = μ). 2) The variance of the series is constant
and does not depend on time (Var(Xₜ) = σ²). 3) The autocovariance between any two
observations depends only on the lag (the time difference) between them, not on the specific
time t (Cov(Xₜ, Xₜ₊ₕ) = γₕ).
,[Type here]
Question 3
Which of the following is a characteristic of a white noise process?
A) The observations are correlated with a constant mean and variance.
B) The observations have a constant mean of zero, a constant variance, and are serially
uncorrelated.
C) The observations have a mean of zero but a changing variance.
D) The observations are serially correlated with a zero mean.
E) The observations follow a normal distribution with a non-zero mean.
Correct Answer: B) The observations have a constant mean of zero, a constant variance,
and are serially uncorrelated.
Rationale: A white noise process is the building block of many time series models. It is a
sequence of random variables with a mean of zero, a constant variance, and zero
autocovariance for all non-zero lags. This means that each observation is independent of all
others.
Question 4
What does the Autocorrelation Function (ACF) measure?
A) The correlation between a time series and a different time series.
B) The correlation between a time series and its own past values at different lags.
C) The correlation between a time series and its future values.
D) The partial correlation between observations after removing the effect of intervening lags.
E) The variance of the time series at different points in time.
Correct Answer: B) The correlation between a time series and its own past values at
different lags.
Rationale: The ACF at lag k, denoted ρ(k), measures the linear relationship between an
observation at time t (Xₜ) and the observation k periods in the past (Xₜ₋ₖ). It is a key tool
for identifying the structure of a time series.
Question 5
What does the Partial Autocorrelation Function (PACF) at lag k measure?
A) The correlation between Xₜ and Xₜ₋ₖ after removing the linear dependence on the
intermediate observations (Xₜ₋₁, Xₜ₋₂, ..., Xₜ₋ₖ₊₁).
,[Type here]
B) The total correlation between Xₜ and Xₜ₋ₖ.
C) The correlation between the first k and the last k observations of a series.
D) A moving average of the autocorrelations up to lag k.
E) The probability that the autocorrelation at lag k is significant.
Correct Answer: A) The correlation between Xₜ and Xₜ₋ₖ after removing the linear
dependence on the intermediate observations (Xₜ₋₁, Xₜ₋₂, ..., Xₜ₋ₖ₊₁).
Rationale: The PACF provides a "cleaner" view of the direct relationship between two
observations by removing the confounding effects of the observations that lie between
them. It is crucial for identifying the order of autoregressive (AR) models.
Question 6
Classical decomposition of a time series typically breaks it down into which four components?
A) Mean, Variance, Skewness, and Kurtosis
B) Trend, Seasonality, Cyclical, and Irregular (Noise)
C) Autoregressive, Moving Average, Integrated, and Seasonal
D) Level, Slope, Period, and Error
E) Past, Present, Future, and Random
Correct Answer: B) Trend, Seasonality, Cyclical, and Irregular (Noise)
Rationale: The classical decomposition model assumes that a time series can be broken
down into these four underlying components: the long-term Trend, the fixed-period
Seasonality, the non-fixed-period Cyclical fluctuations, and the random, unpredictable
Irregular component.
Question 7
What is the primary purpose of using a moving average to smooth a time series?
A) To make the series stationary.
B) To forecast future values with perfect accuracy.
C) To highlight the underlying trend and remove short-term fluctuations or noise.
D) To identify the order of an ARMA model.
E) To calculate the partial autocorrelations of the series.
, [Type here]
Correct Answer: C) To highlight the underlying trend and remove short-term fluctuations
or noise.
Rationale: A moving average works by replacing each data point with the average of itself
and a certain number of its neighbors. This averaging process smooths out the random,
short-term "jumps" in the data, making the longer-term trend or pattern more visible.
Question 8
A time series analyst observes an Autocorrelation Function (ACF) plot that shows a gradual,
exponential decay, and a Partial Autocorrelation Function (PACF) plot that cuts off abruptly after
lag 1. What is the most likely model for this time series?
A) An MA(1) model
B) An AR(1) model
C) An ARMA(1,1) model
D) A white noise process
E) An MA(2) model
Correct Answer: B) An AR(1) model
Rationale: This is the classic "signature" of an Autoregressive model of order 1 (AR(1)). An
AR(p) process is characterized by a PACF that cuts off after lag p and an ACF that tails off
(decays exponentially or with a sinusoidal pattern).
Question 9
A time series analyst observes an Autocorrelation Function (ACF) plot that cuts off abruptly after
lag 2. The Partial Autocorrelation Function (PACF) plot appears to tail off exponentially. What is
the most likely model for this time series?
A) An AR(2) model
B) An ARMA(2,2) model
C) An MA(2) model
D) An ARIMA(2,1,0) model
E) A white noise process
Correct Answer: C) An MA(2) model
Rationale: This is the classic "signature" of a Moving Average model of order 2 (MA(2)).
