STATS 326代做、R设计代做、data留学生代写、R编程语言调试

日期：2020-03-13 09:55

STATS 326

Applied Time Series

ASSIGNMENT ONE

Due: 20 March, 12.00

Worth 6% of your final grade (727 4%)

Hand-in to the appropriate STATS 326 Hand-in box in the Student Resource Centre

For each of the first 4 questions in this assignment you are required to find a Univariate Time

Series, describe the Time Series you are using, state where you found the data (web address is

fine), do a Time Series plot of the data in R (see page 14 of the Course Notes) and describe

the main features you see in the plot.

Each Time Series you use can come from the web, journal articles, text books etc. You must

NOT use any Time Series that are used as examples in the course (see page v of your Course

Notes for a list of the Time Series data sets that are used as examples in the Lecture Notes

and page 243 for a list of the Tutorial data sets). Using a Time Series used in the course with

the time range changed (reduced or increased) is not acceptable. It would also be advisable to

look at pages 55 – 59 and pages 244 – 248 before commencing this assignment.

This assignment will be marked out of 100 with each question worth 20 marks. Each question

should take no more than 1 side of A4 paper. You are encouraged to print your assignment

“2-up” to save paper.

Question One: [20 marks]

Find a Time Series that exhibits cycles. (See pages 55 – 58 of the Course Notes.)

Question Two: [20 marks]

Find a Stationary Time Series. (See page 3 of the Course Notes.)

Question Three: [20 marks]

Find a Time Series that has a seasonal component but no trend or cycle. (See pages 55 – 59 of

the Course Notes.)

Question Four: [20 marks]

Find a Time Series that has a reasonably linear trend and a seasonal component. (See page 55

and pages 58 – 59 of the Course Notes.)

Question Five: [20 marks]

The data contained in the file “Cape Grim CO2 2000.1 - 2019.9.txt” records the average

concentration of CO2 in the atmosphere at Cape Grim, Tasmania, Australia for each month

from January 2000 to September 2019.

Load the data into R, create a “time series object” and produce a Time Series plot. Copy the

plot into your assignment.

Using the aggregate function in R, convert the data into the average concentration of CO2

in the atmosphere for each quarter from 2000 to 2019.3. (Include the R commands you used

to aggregate the data in your assignment.) Plot the quarterly series, copy the plot into your

assignment and describe the plot.

Data Source:

https://www.esrl.noaa.gov/gmd/dv/data/index.php?category=Greenhouse%2BGases&parame

ter_name=Carbon%2BDioxide&frequency=Monthly%2BAverages&site=CGO

HINT: See page 233 of the Course Notes. Here we aggregate monthly data into annual data.

The original frequency was 12 and the new frequency is 1 (in R: nfrequency = 1

which is the default setting so was not required as an argument in the R command).

Our data for this question is monthly (frequency = 12). The new frequency is 4 (in

R: nfrequency = 4).

NOTE: Check that the quarterly averages appear consistent with the original monthly

averages.