Graded Quiz 4
pdf
keyboard_arrow_up
School
Arizona State University *
*We aren’t endorsed by this school
Course
578
Subject
Industrial Engineering
Date
Feb 20, 2024
Type
Pages
7
Uploaded by BrigadierMusicWolverine27
Question 1 What are characteristics of the Autoregressive Time Series Model, AR(2)? Select all that apply. •
The model depends on a random component contributor. •
The model has a closed form formula. •
The model depends on the previous 2 time instances of past value. •
The model depends on the immediate past value. Question 2 Which conditions are necessary
for stationary time series? Select all that apply. •
The time series must have a smooth shape without too much fluctuation in amplitude. •
Mean must be constant and does not depend on time. •
The autocovariance function must depend on s and t only through their difference |s-t| (where t and s are moments in time). •
The time series under considerations must have a stationary variance over time. Flag question: Question 3 Question 3 Which graph (or set of graphs) can be used to detect seasonality in time series data? •
Multiple box only •
Autocorrelation only •
Line graph only •
Multiple box and autocorrelation Question 4
Figure 1: Time Series A[t] and B[t]
Figure 2: Matrices
Review Figure 1: Time Series A[t] and B[t]
, and Figure 2: Matrices
. For time series and , the distance between them is defined by. Which matrix represents the distance matrix between the two time series? •
Matrix 3 •
Matrix 1 •
Matrix 2 Question 5 In terms of the dynamic programming implementation of DTW algorithm presented in this course, which moves are the kinds that the dynamic time warping algorithm assumes when moving through the cells of the matrix in finding the best path? Select all that apply. •
Down •
Up •
Right •
Up-and-Right Question 6 Which property is desirable for a time series process? •
Variability •
Stationarity
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help
•
Independence of variables •
Identical distribution of variables Question 7 Table: Symbol Map
Range Symbol [0, 0.5) A [0.5, 1) B [1, 1.5) C [0, -0.5) D [-.5, -1) E [-1, -1.5) F Figure: Time Series
Review Table: Symbol Map
, and Figure: Time Series
. The table is a symbol map of a range of values. In terms of Symbolic Aggregate Approximation (SAX) algorithm, what is the symbolic representation of the provided time series? •
AAABCFED •
FAAAEDBC •
FEDAAABC •
FEBCDAAA Question 8 In forecasting with time series analysis, suppose that the demand for a component is 100 in October 2016, 200 in November 2016, 300 in December 2016, and 400 in January 2017. What is the second 3-month simple moving average? •
300 •
Need more information •
250 •
200 Question 9 Figure: Observations
Review Figure: Observations
. Which observations have a small
correlation-based distance between them? •
Observation 1 and Observation 3 •
Observation 2 and Observation 3 •
Observation 1 and Observation 2 Question 10 Imagine that you are working on a time series dataset. Your manager has asked you to build a highly accurate model for model-based time series analysis. Now you are considering building one of two types of model: Model Type 1: Moving Average Model (MA) Model Type 2: Autoregressive Model (AR) In this real-world scenario, what is your rationale when choosing which model to use to find the most accurate solution?
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help
•
Model 2 is always better than Model 1 because it is able to capture the past observations at different lags. •
Both Model 1 and Model 2 are too simple and we should always combine them together to make a complex ARMA or ARIMA model but not use either one of them alone. •
Model 1 is always better than Model 2 because it captures the error from external events at different lags in the past. •
Depending on the application and data, either Model 1 or Model 2 can be helpful for the analysis. We should consider the tradeoff of complexity and fit.