Unit 5 - Practice Quiz

INT395

1 Which of the following best describes a Time Series dataset?

A. A collection of data points collected from random individuals at a single point in time.
B. A sequence of data points indexed in time order, often with equal intervals.
C. A dataset where the target variable is categorical and independent of time.
D. A dataset containing only images captured at different locations.

2 In the context of Time Series analysis, what does Stationarity imply?

A. The time series has a constant trend increasing over time.
B. The statistical properties (mean, variance, covariance) are constant over time.
C. The data contains no missing values.
D. The seasonal component of the data is strictly periodic.

3 Which component of a time series represents a long-term increase or decrease in the data?

A. Seasonality
B. Noise (Irregularity)
C. Trend
D. Cyclicality

4 What is the difference between Univariate and Multivariate time series?

A. Univariate has irregular intervals; Multivariate has regular intervals.
B. Univariate involves a single variable varying over time; Multivariate involves multiple variables varying over time.
C. Univariate uses regression; Multivariate uses classification.
D. Univariate is for stationary data; Multivariate is for non-stationary data.

5 When preparing time series data for supervised learning, what is a Lag Feature?

A. The difference between the current value and the mean.
B. A future value used to predict the past.
C. Observation from a previous time step used as an input feature (e.g., ).
D. The error term from a moving average model.

6 Which of the following is a common method to handle Irregular Intervals in time series data before modeling?

A. Dropping all timestamps.
B. Resampling or Interpolation to a fixed frequency.
C. Using a random number generator to fill gaps.
D. Ignoring the time index and treating it as cross-sectional data.

7 Why is K-Fold Cross-Validation with random shuffling inappropriate for time series data?

A. It is too computationally expensive.
B. It causes data leakage by using future data to predict past data.
C. Time series data is always too small for K-Fold.
D. It assumes the data is non-stationary.

8 What is the Walk-Forward Validation (or Time Series Split) strategy?

A. Randomly selecting 20% of the data for testing.
B. Training on all data and testing on the same data.
C. Using a sliding or expanding window where the training set consists of past data and the test set consists of immediate future data.
D. Splitting data based on the magnitude of the values.

9 Which mathematical operation is commonly used to transform a non-stationary series with a trend into a stationary one?

A. Integration
B. Differencing (e.g., )
C. Multiplication
D. Squaring

10 In an Autoregressive (AR) model of order , denoted as AR(), the current value is predicted based on:

A. Past forecast errors.
B. A linear combination of past values.
C. A moving average of the entire dataset.
D. Independent external variables only.

11 The equation represents which model?

A. AR(1)
B. MA(1)
C. ARIMA(1,1,1)
D. Random Walk

12 What does the 'I' in ARIMA stand for?

A. Independent
B. Interpolated
C. Integrated
D. Irregular

13 In the notation ARIMA(p, d, q), what does the parameter represent?

A. The number of lag observations included in the model.
B. The size of the moving average window.
C. The number of times that the raw observations are differenced.
D. The seasonality period.

14 Which plot is most useful for determining the order of a Moving Average (MA) model?

A. Scatter plot
B. Histogram
C. Partial Autocorrelation Function (PACF)
D. Autocorrelation Function (ACF)

15 Which plot is most useful for determining the order of an Autoregressive (AR) model?

A. Autocorrelation Function (ACF)
B. Partial Autocorrelation Function (PACF)
C. Box Plot
D. Spectral Density Plot

16 What distinguishes SARIMA from ARIMA?

A. SARIMA handles missing values better.
B. SARIMA explicitly models Seasonal components.
C. SARIMA is only for multivariate data.
D. SARIMA does not require stationarity.

17 In SARIMA notation , what does represent?

A. The maximum value of the series.
B. The number of features.
C. The number of time steps for a single seasonal period.
D. The order of the polynomial trend.

18 If a time series is a White Noise process, which of the following is true?

A. It can be perfectly predicted using linear regression.
B. Its mean and variance change rapidly over time.
C. It has zero autocorrelation for all lags .
D. It strictly follows a sinusoidal wave.

19 What is the primary risk of using standard Min-Max Scaling on the entire dataset before splitting into train/test in time series?

A. It makes the data non-stationary.
B. It introduces Lookahead Bias (Data Leakage).
C. It changes the time order.
D. It removes the seasonality.

20 Which test is commonly used to check for Stationarity?

A. T-test
B. Augmented Dickey-Fuller (ADF) test
C. Chi-Square test
D. ANOVA

21 Consider the equation: . This is an example of:

A. AR(1) with unit root.
B. Stationary AR(1) process.
C. Explosive AR(1) process.
D. MA(1) process.

22 In the context of Time Series Regression, what are Exogenous Variables?

A. Variables that are predicted by the model.
B. Internal lag variables derived from the target.
C. External variables that influence the target but are not influenced by it.
D. Variables representing the error term.

23 If a time series exhibits a variance that increases over time (heteroscedasticity), which transformation is commonly applied to stabilize variance?

A. Differencing
B. Logarithmic Transformation
C. Adding a constant
D. Standardization

24 What is an Expanding Window split strategy?

A. The training set size remains constant and slides forward.
B. The training set starts small and grows in size as new data becomes available.
C. The test set size grows while the training set shrinks.
D. Data is shuffled and split randomly.

25 What is the consequence of Over-differencing a time series?

A. The series becomes perfectly linear.
B. It introduces artificial noise and negative autocorrelation.
C. It makes the trend more pronounced.
D. It increases the variance of the trend.

26 Which of the following describes a Random Walk model?

A. ARIMA(0, 1, 0)
B. ARIMA(1, 0, 0)
C. ARIMA(0, 0, 1)
D. ARIMA(1, 1, 1)

27 In an AR(1) model , if , the series is:

A. Stationary
B. Mean reverting
C. A Random Walk (Non-stationary)
D. White Noise

28 Which metric provides a penalty for adding parameters to a model, helping to avoid overfitting in ARIMA model selection?

A. MSE (Mean Squared Error)
B. R-Squared
C. AIC (Akaike Information Criterion)
D. MAE (Mean Absolute Error)

29 For a monthly time series with strong yearly seasonality, what is the value of the seasonal period parameter ?

A. 4
B. 7
C. 12
D. 30

30 What is Seasonal Differencing?

A.
B. where is the seasonal period.
C.
D.

31 A model defined as is an:

A. AR(1)
B. AR(2)
C. MA(2)
D. ARMA(1,1)

32 The Ljung-Box Test is applied to the residuals of a time series model to check:

A. If the residuals are normally distributed.
B. If the residuals differ significantly from white noise (i.e., if there is remaining autocorrelation).
C. If the model parameters are statistically significant.
D. If the original data was stationary.

33 How are missing values typically handled in time series if dropping them is not an option?

A. Replace with zero.
B. Time-based interpolation (e.g., linear) or Forward Fill.
C. Replace with the dataset maximum.
D. Replace with the dataset standard deviation.

34 Which component is represented by in SARIMA?

A. Seasonal Autoregressive order
B. Seasonal Moving Average order
C. Seasonal Differencing order
D. Trend Differencing order

35 If the ACF plot decays slowly (linear decay) while the PACF shows a sharp cutoff at lag 1, the series is likely:

A. MA(1)
B. AR(1)
C. White Noise
D. ARMA(1,1)

36 In time series regression, what is a Rolling Window feature?

A. A feature computed by aggregating (mean, max, etc.) values over a fixed-size window of past observations.
B. A window that randomly selects features.
C. A feature that looks into the future.
D. A technique to increase the sampling rate.

37 Which of the following equations represents an ARIMA(1,1,1) model (excluding constant)? Let .

A.
B.
C.
D.

38 When encoding cyclical time features like 'Month of Year' (1-12) for regression, what is the best approach to preserve the cycle (i.e., Dec is close to Jan)?

A. Label Encoding (1, 2, ..., 12)
B. One-Hot Encoding
C. Sine and Cosine transformation
D. Binary Encoding

39 What is Naive Forecasting?

A. Using a complex neural network.
B. Predicting the current value as the average of all past values.
C. Predicting the value at time to be equal to the value at time .
D. Predicting zero for all future steps.

40 What is the key assumption of the Moving Average (MA) model regarding shocks/errors?

A. Current value depends on past values of the variable.
B. Current value depends on current and past unobserved white noise error terms.
C. Errors are correlated with the independent variables.
D. Errors increase exponentially with time.

41 In the context of SARIMA, what is an Exogenous regressor often denoted as?

A. SARIMAX
B. ARIMAS
C. ARIMAE
D. XY-ARIMA

42 If a time series has a unit root, the series is:

A. Stationary
B. Non-Stationary
C. Seasonal
D. Deterministic

43 Which ARIMA model is theoretically equivalent to Simple Exponential Smoothing?

A. ARIMA(0,1,1)
B. ARIMA(1,0,0)
C. ARIMA(2,2,2)
D. ARIMA(0,0,0)

44 What is the purpose of Auto-Regressive terms in a model?

A. To model the relationship between an observation and a lagged observation.
B. To model the relationship between an observation and a residual error.
C. To remove seasonality.
D. To integrate the time series.

45 Why is it important to examine the Residuals of a fitted time series model?

A. To ensure they follow a clear trend.
B. To ensure they have high autocorrelation.
C. To ensure they resemble random white noise.
D. To ensure they are all positive.

46 What implies that a Time Series is Deterministic?

A. It can be perfectly predicted by a mathematical function of time (e.g., ).
B. It contains a stochastic component.
C. It is a random walk.
D. It has infinite variance.

47 In a Multi-step forecast strategy, what is the 'Direct' method?

A. Ideally predicting one step ahead and using that prediction for the next step.
B. Building a separate model for each future time step to be predicted.
C. Using only the most recent observation.
D. Averaging the results of AR and MA models.

48 What is the 'Recursive' (or Iterative) multi-step forecast strategy?

A. Predicting , then using that prediction as input to predict , and so on.
B. Training on future data.
C. Using a different algorithm for every step.
D. Predicting all steps simultaneously using a vector output.

49 Which transformation is the inverse of Differencing?

A. Differentiation
B. Cumulative Sum (Integration)
C. Logarithm
D. Exponentiation

50 When is an ARIMA model considered Invertible?

A. When the AR polynomial roots lie inside the unit circle.
B. When the MA polynomial roots lie outside the unit circle.
C. When it can be rewritten as an infinite AR process.
D. When .