MINITAB1

1

This question relates to the data in the Unemployment rate.csv workbook

(a)(i) For these data, decide on the appropriate visualisation of the two unemployment rate series – where the goal is to have both series on the same set of axes.

Explain in less than 300 words why you chose this approach.

(b)(i) Take the unadjusted (original) unemployment data. Use Minitab to undertake a time series decomposition of the data.

Using an additive decomposition, provide the following:

⦁ The graph of actual, fit and trend.
⦁ Up to 300 words to explain the following: What does the gap between the actual and the fit represent?

(c)(i) Is there a seasonal pattern? Describe this pattern. Does it match your intuition and why?

(ii) Is there a positive and negative trend in the data?

In less than 300 words explain why this trend may be due to omitted variable bias rather than a real time trend, and what this might mean for forecasting.

2

This question relates to the data in the Time_use.csv workbook

(a)(i) In class we created a scatterplot of the sleep data. Provide a scatterplot of the sample work length data.

(b)(i) Use Minitab or a programme of your choosing to complete the following table for the sleep and work data. Write your answers as accurately as you can to one decimal place in the table below if they are not an integer.

Statistic Sleep sample Work sample
Mean 6.9983 11.973
Median 6.9922 11.946
Minimum 4.3083 1.534
Maximum 9.6934 23.276
Standard error of the mean 0.00699 0.0300
Interquartile range Q1 65323
Q2 7.4715 Q1 9.961
Q2 14.015
Standard deviation 0.6990 3.005
Range 4.3083-9.6934 1.534-23.276
Skewness 0.0082004 0.0231497

(ii) In less than 300 words, explain why there might be measurement error in this data – and any indication of this you may have seen in the data. [Hint: Consider what the data is, how people would respond to it, and what is observed.]

3

This question relates to the data in the NSW population.xlsx workbook

(a)(i) Choose an appropriate visualisation to compare the age groups by sex – specifically we want to understand the sex ratios for our age groups. Explain why you have picked this approach. Hint: You will want to look at “values from a table” rather than counts for this approach.


(b)(i) If we were instead thinking about a “population pyramid” (an example is available in that hyperlink), are there problems with the groupings that are given in data? [Note: You do not need to create the pyramid.]

(ii) If this was all the information that we had, how could you adjust the data in order to consider a population pyramid. [Note: You do not need to create the pyramid, describe in words.]

4

Provide the link to, or a copy and pasted image of, a graph you believe is a “misleading graph” from online.

Explain why it is misleading in 300 words or less using the criteria from lectures.