Deck 17: Seasonal Variation and Forecasting

Question 1

Using the additive model for time series forecasting, estimate the the value of y when t = 5.2 (year 5 quarter 2). In year 1, quarter 1 the value of t is 20 (with an upward trend of 5 per quarter). s = +23 for quarter 2.

A) 113
B) 118
C) 123
D) 128

Answer 1

123

Question 2

Which of the following statements concerning forecasting is NOT true?

A) In some cases, simple estimation of likely future trends is sufficient
B) Structured forecasts, using statistics, can be relied upon to be accurate
C) Forecasting allows us to estimate future performance based on historical trends
D) Before making a large investment, structured forecasting may be necessary

Answer 2

Structured forecasts, using statistics, can be relied upon to be accurate

Question 3

In which of the following situations will the forecast be MOST accurate?

A) When the residual factor is large
B) When the residual factor is small
C) When the residual factor equals the seasonal factor
D) When the residual factor is equal to the mean of the time series data

Answer 3

When the residual factor is small

Question 4

In the multiplicative model, seasonal adjustment is made by dividing the value of y by S. Which of the following options is the correct formula for S?

A) S = 1 - (y-t)/t
B) S = 1 + (y+t)/t
C) S = 1+ (y-t)/t

Question 5

Which of the following statements is TRUE?

A) Residual variation cannot be determined for time series data
B) Residual variation may only be calculated for time series data when more than 10 data points exist
C) Residual variation is usually insignificant and so calculating it has little value in practice
D) Residual variation can be calculated when trend values and seasonal variation are known

Question 6

Which of the following options BEST describes seasonal variation?

A) The influence of short-term cyclic variation
B) The influence of long-term cyclic variation
C) The influence of a long-term upwards trend in the data
D) The influence of a long-term downwards trend in the data

Question 7

In which of the following situations would seasonal variation be LEAST important?

A) Predicting the sale of beachwear during a year
B) Predicting the demand for gas over the course of a year
C) Predicting demand for ski holidays in the European Alps
D) Predicting changes to the Hong Kong Stock Exchange index

Question 8

Which value of s indicates that seasonal variation for the given time period deflates the trend by 6?

A) +6
B) -6
C) +0.6
D) -0.6

Question 9

Which option BEST describes the purpose of seasonally adjusting time series data?

A) Adjusting data to give a picture of data 'as if seasons had not existed'
B) Adjusting data so that the seasonal variation is constant
C) Adjusting data so that the seasonal variation is reduced by half
D) Adjusting data to show the impact of seasonal differences

Question 10

The range of trend factors for data indicating the sales of French wine (in millions of litres) are 2.4 to 3.8 - there are 6 trend values. Assuming a value of S of 0.65, estimate the next value in the sequence (2 decimal places).

A) 2.29
B) 2.59
C) 3.52
D) 3.98