Instructions
STAT40180
Data Programming with R
Assignment 2
This assignment is due on Friday 8th November 2019 at 5pm.
You should submit it to the Assignment 2 assignment object in Brightspace.
You should submit two files only:
1. Rmd file detailing the commented code you used to obtain your answers.
2. finaldocumentineitherpdforWordwhichshouldcontainanswerstothequestions
below.
If you created an HTML file, please convert it to pdf. You can use Google Chrome: File > Print > Destination [Change] > select Save as PDF.
You may submit it multiple times before the deadline, but only the last version will be marked.
There is a maximum of 10 marks for this assignment. This assignment is worth 10% of your final grade.
The marks available for each question are shown in brackets.
Late submissions will score 0, unless a Late Submission of Coursework form is sub-
mitted.
You may have to discover and learn some new functions. Use help() and help.search() to find what you need.
Some tips on using R Markdown are given at the end of this document.
1
The file mly532.csv contains the Historical Monthly Data recorded at the Dublin Airport Met E ireann Weather Observing Station from November 1941 to May 20191. The data set contains the following variables:
year Year month Month
rain Precipitation Amount (mm) maxtp Maximum Air Temperature (C) mintp Minimum Air Temperature (C) mnmax Mean Maximum Temperature (C) mnmin Mean Minimum Temperature (C)
gmin Grass Minimum Temperature (C)
wdsp Mean Wind Speed (knot) maxgt Highest Gust (knot)
sun Sunshine duration (hours)
1. Complete your assignment using R Markdown, check that all the output and code are correctly shown in your final document. Knit your document frequently to fix errors. Once completed, submit the Rmd file and the resulting pdf or word document which shows all your code. [0.5]
2. Load in the data as an object called DublinAirport. Notice that you have to skip the first 19 lines contained in the file before beginning to read data. Line 20 contains the column names. Display the structure of the dataset. [0.5]
3. Transform the column months to a factor with labels given by the month names. [Two hints: (1) look at the arguments levels and labels of the function factor. (2) you may want to make use of the built-in constant month.name] [1]
4. Use the aggregate function to compute which month has on average the highest and the lowest Precipitation Amount. [1]
5. Create a new column which contains a factor indicating the season:
Winter: December, January, February,
Spring: March, April, May,
Summer: June, July, August,
Autumn: September, October, November
[Hint: again, look at the arguments levels and labels of the function factor.] [1]
6. Assign to the DublinAiport object the classes WeatherData and data.frame. [0.5]
1Source: https://www.met.ie/climate/available-data/historical-data
2
7. Write an S3 summary method for an object of class WeatherData which produces the following statistical summaries the rain, maxtp, mintp, maxgt variables split by season: mean, standard deviation, minimum, maximum. Ignore the missing values in the calculations.
Test your function on the DublinAirport data set and comment your findings. [2.5]
8. Create an S3 plot method for the class WeatherData that produces the following plots.
(1) Plot of the monthly Air Temperature (C) (maxtp, mintp).
(2) Plot of the Precipitation Amount (mm) (rain).
(3) Plot of the Highest Gust (knot) (maxgt).
The user must be able to decide which years to plot. By default it will use the data from 2015 until 2018.
The user must be able to decide which plot to draw (i.e, only one of the three, two of the three, or all three plots). By default the function will create all three plots.
The plots must be on a single panel.
The plots must have meaningful labels and/or titles, and a legend if needed.
Test your function on the DublinAirport data set.
[3]
3
Tips for R Markdown
Be aware that a common error is to give the same label to two different code chunks!
`{r cars}
summary(cars)
`
`{r cars}
plot(cars)
`
You can fix this by changing the label to one of them:
`{r cars2}
plot(cars)
`
If you want to improve the appearance of your plot in your knitted document you can set up the dimension of your figure:
`{r, fig.height = 10, fig.width = 7, fig.align = center}
plot(Nile)
`
In case of an error in your code, add the option error = TRUE into the R chunk to run the code, show the error message on the knitted file. For example:
`{r, error = TRUE}
x <- “a”sum(a)“` For all the available options for the R chunk, you can see here: https://yihui.name/ knitr/options/ R Markdown website: https://rmarkdown.rstudio.com/ R Markdown cheatsheet is available here: https://www.rstudio.com/resources/ cheatsheets/#rmarkdown4
Reviews
There are no reviews yet.