Business Forecasting Analytics – ADM 4307 Sample Final Exam – Fall 2019
Duration: 2 hours and 30 minutes
Professor:
Last Name: ________________________ First Name: _____________________
Copyright By PowCoder代写 加微信 powcoder
Student #: ________________________
Instructions:
1- Write in your last and first name and your Student ID number in the spaces above, and sign
the Statement of Academic Integrity below.
2- Verify that your exam copy has 8 pages (including this title page). If yours does not, please
inform the professor now.
3- Please make sure you record all questions in Part A on the provided answer sheet (scantron).
Only answers in this answer sheet will be marked.
4- This is a closed-book exam: however, one double sided formula-sheet (8.5″ x 11″) and a
pocket calculator are allowed. The formula-sheet must be handed over with the exam
copy at the end of the exam period, or else your copy will not be marked.
5- Read each question very carefully and do not hesitate to ask for clarifications when needed.
6- NO COMMUNICATION DEVICES MAY BE WITHIN SIGHT.
7- Good luck and keep smiling.
DO NOT WRITE ON THE TABLE BELOW
1-50 Multiple Choice Questions 1 x 50 = 50 use scantron 51-55 Short Answer Questions 10 x 5 = 50 use this booklet
Description
Answer Sheet
Statement of Academic Integrity
The School of Management does not condone academic fraud, an act by a student that may result in a false academic evaluation of that student or of another student. Without limiting the generality of this definition, academic fraud occurs when a student commits any of the following offences: plagiarism or cheating of any kind, use of books, notes, mathematical tables, dictionaries or other study aid unless an explicit written note to the contrary appears on the exam, to have in his/her possession cameras, radios (radios with headsets), tape recorders, pagers, cell phones, or any other communication device which has not been previously authorized in writing.
Statement to be signed by the student:
I have read the text on academic integrity and I pledge not to have committed or attempted to commit academic fraud in this examination.
Signed:______________________________________
Note: an examination copy or booklet without that signed statement will not be graded and will receive a final exam grade of zero.
ADM 4307 Sample Final Exam Fall 2019
PART A – True/False and Multiple-Choice Questions
Read each question carefully, and then CIRCLE THE ANSWER that best fits the question.
1. Forecasting techniques generally assume that the same causal system that existed in the past will continue to exist in the future.
2. The primary difference between seasonality and cycles is:
A. the duration of the repeating patterns.
B. the magnitude of the variation.
C. the ability to attribute the pattern to a cause.
D. There is more forecasting “noise” in a cycle.
E. There is less forecasting “noise” in a cycle.
3. Sales for a product have been consistent over several years, although showing a steady
upward trend. The company wants to understand what drives sales. The best forecasting technique would be:
A. trend models.
B. judgmental methods.
C. moving averages.
D. regression models.
E. exponential smoothing techniques
4. A company is conducting long-term planning of which types of services they should offer. Which of the following forecasting techniques are they most likely to use?
A. Trend models
B. Executive opinion
C. Regression models
D. Simple exponential smoothing
E. Naïve method
5. Visualization technique that summarizes the relationships of several pairs of variables A. Box plots B. Histogram
C. Scatter plots D. Contour plots
E. None of above
6. Forecast accuracy is based on what?
A. Test set
B. Training set
D. Neither
7. Which of the following is not a component of time series decomposition?
A. Seasonal
B. Trend-Cycle
C. Autocorrelation
D. Remainder
8. Data that exhibits rises and falls that are not of a fixed period is known as what?
B. Seasonal
C. Cyclical
D. White Noise
9. What does autocovariance measure?
A. Linear dependence between multiple points on the different series observed at different
B. Quadratic dependence between two points on the same series observed at different
C. Linear dependence between two points on different series observed at same time
D. Linear dependence between two points on the same series observed at different times
Sample Final Exam
Refer to Figure 11. Line A is which simple forecasting method?
A. Average Method
B. Naïve Method
C. Seasonal Naïve Method
D. Drift Method
Refer to Figure 1. Line B is which simple forecasting method?
A. Average Method
B. Naïve Method
C. Seasonal Naïve Method
D. Drift Method
Refer to Figure 1. Line C is which simple forecasting method? A. Average Method
B. Naïve Method
C. Seasonal Naïve Method
D. Drift Method
Noise objects are always considered as outliers. A. TRUE
Line B (Straight Line)
(Not Straight)
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
10 6 8 12 11 7 9 13
14. Refer to Figure 2. Using the average method, what is the forecast of Quarter 2 of Year 3? A. 7
15. Refer to Figure 2. Using the naïve method, what is the forecast of Quarter 2 of Year 3? A. 7
16. Refer to Figure 2. Using the seasonal naïve method, what is the forecast of Quarter 2 of Year
17. Refer to Figure 2. Using the drift method, what is the forecast of Quarter 2 of Year 3? A. 7
ADM 4307 Sample Final Exam Fall 2019
18. Which of the following is a data quality problem?
A. Noise and outliers
B. Missing values
C. Duplicate data
D. Wrong data
E. All of the above
19. Which of the following is relatively easier to estimate in time series modeling?
A. Seasonality
B. Cyclical
C. No difference between Seasonality and Cyclical
20. Neural network training is accomplished by repeatedly passing the training data through the
network while
A. individual network weights are modified
B. training instance attribute values are modified
C. the ordering of the training instances is modified
D. individual network nodes have the coefficients on their corresponding functional
parameters modified
E. all of the above
21. Stationarity is a desirable property for a time series process.
22. Which of the following circumstances is likely to make a forecast using (multiple) regression
analysis less reliable?
A. All the points lie exactly along the regression line in the scatter diagram
B. All the relevant variables are included in the regression equation
C. No important variables are missing from the regression equation
D. Some important variables are missing from the regression equation
23. Seasonal variation can be estimated using dummy variables in linear regression analysis.
24. The greater the number of periods used to calculate a moving average, the more sensitive the
forecast is to the most recent observation.
25. When we use an approach which implies that the forecast for the next time period should take into account the observed error in the earlier forecast for the current time period, then we are using:
A. Regression analysis
B. Time series analysis
C. ARIMA method
D. Exponential smoothing method
26. The Delphi method generates forecasts by surveying consumers to determine their opinions.
27. The first step in time-series analysis is to
A. perform preliminary regression calculations. B. calculate a moving average.
C. plot the data on a graph
D. identify relevant correlated variables
28. In performing diagnostics for an ARMA(1,1) model fit, I see the following output in R:
> runs(rstandard(data.arma11.fit))
How do I interpret this output?
A. The standardized residuals seem to be well modeled by a normal distribution.
B. The standardized residuals are not well represented by a normal distribution.
C. The standardized residuals appear to be independent.
D. We should probably consider a model with either p > 1 or q > 1 (or both).
ADM 4307 Sample Final Exam Fall 2019
#> Series: h02
#> ARIMA(3,0,1)(0,1,2)[12]
#> Box Cox transformation: lambda= 0
#> Coefficients:
#> ar1 ar2 ar3 ma1 sma1 sma2 #> -0.160 0.548 0.568 0.383 -0.522 -0.177 #> s.e. 0.164 0.088 0.094 0.190 0.086 0.087 #>
#> sigma^2 estimated as 0.00428: log likelihood=250 #> AIC=-486.1 AICc=-485.5 BIC=-463.3
#> Ljung-Box test
#> data: Residuals from ARIMA(3,0,1)(0,1,2)[12] #> Q* = 51, df = 30, p-value = 0.01
#> Model df: 6. Total lags used: 36
29. Based on the R output above, the ARIMA model passes the residual test.
30. Based on the R output above, the ARIMA model needs a first-order differencing.
1950 1952 1954
1956 1958 1960
31. Based on plot in Figure 4, we need a transformation for this dataset to develop a easier and more accurate model.
32. During lecture, I always told you that motto to comment on the precision of her predictions about my future. What is that motto?
A. All models are wrong, but some are useful
B. All models are perfect and useful
C. None of the models are perfect, but some are useful
D. All models are useful
33. Based on residual test in Figure 5, which of the following is not true
A. The model is satisfactory.
B. No obvious patterns are visible on the graph of the residuals
C. Ljung-Box test suggests that the residuals are white noise
D. The residuals look close enough to normally distributed
E. None of the above
34. Based on the R output in Figure 5, the ARIMA model needs a seasonal differencing.
ADM 4307 Sample Final Exam Fall 2019
ADM 4307 Sample Final Exam Fall 2019
PART B – Short Answer Questions
51. Define forecasting? List and shortly explain the assumptions and the features common to all forecasts.
52. What are the basic steps in a forecasting task. Explain them shortly.
53. Explain the main idea behind the time series and associative models? What are the main
differences between them? In case of a close accuracy results, which one do you prefer to use for forecasting? Why?
54. Define the residual and forecasting error? List and shortly explain the key differences between them?
55. Define Qualitative Forecasting. When do we need Qualitative Forecasting? List and shortly explain three methods.
56. Define mathematical transformation and explain the reason why we need this. How do you understand the requirement for a mathematical transformation?
57. Define time series decomposition and explain the reason why we need this.
58. After selecting the regression variables and fitting a regression model, it is necessary to
evaluate the regression model. How do we evaluate a regression model?
59. Explain the ‘ Method shortly.
60. What are the elements of a good forecast? List and explain 3 of them.
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com