732A94 Advanced R Programming

Computer lab 2

Krzysztof Bartoszek, Bayu Brahmantio, Simon Jorstedt, Duc Tran

(designed by Leif Jonsson and Måns Magnusson)
(converted to LATEX by Arash Haratian)

11 September 2024 (SU10,SU11,SU24,SU25)

Lab deadline: 22 September 23:59

 Instructions

• This lab should be done individually

• It is OK to discuss the problems with each other, but to copy code is NOT allowed. Your solutions
will be checked through URKUND.
• Copying solutions of others and from any online or offline resources is NOT allowed.

• Significant collaborations/discussions should be acknowledged in the solution file.

• The lab should be turned in through LISAM as a documented .R file. In case of problems or if you
do not have access to LISAM it may be emailed to [email protected] or [email protected]
or [email protected] or [email protected].

• In the .R file two text element (objects) are mandatory:

– name (your name)

– liuid (your LiU-ID). If you do not have one yet, please use a dummy one: nnnnn000.

• The deadline for the lab is on the lab’s title page.

• The lab will be graded as follows. Your source file will be downloaded and R will best started with
an empty workspace. Then, the following R commands will be executed by the person grading.
> library(markmyassignment)
> lab path <- "path to the lab’s .yml file"
> set assignment(lab path)
> source("your file.R")
> mark my assignment()
Be careful that in your handed-in file you do not have any lines of code apart from the bodies of
the required functions, liuid and name. Any extras can easily break the above workflow, resulting
in returns.
Depending on the version of markmyassignmentthere might be some issues, when it is requested
to check specific tasks. It will sometimes also check additional tasks, and if they werenot in the
workspace, errors are raised. Please only consider the comments specific to the task you are inter-
ested in.
• Unless explicitely instructed, you may not use any global variables. Using them will result in an
immediate failure of your code.If at any stage your code changes any options, these changes have
to be reverted before your code finishes.

1
LINKÖPING UNIVERSITY
STIMA
Department of Computer and Information Science Advanced R Programming

Contents

1 Lab assignments 4
1.1 Conditional statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1.1 sheldon game(player1, player2) . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 for loops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.1 my moving median() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.2 for mult table() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.3 * cor matrix() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 while loops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.1 find cumsum() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.2 while mult table() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.3 * trial division factorization() . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4 repeat and loop controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4.1 repeat find cumsum() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4.2 repeat my moving median() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.5 Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.5.1 in environment() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.5.2 * where() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.6 Functionals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.6.1 cov() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.7 Closures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.7.1 moment() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.7.2 * mcmc counter factory() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Automatic feedback with markmyassignment
As a complement to get fast feedback on your lab assignments the package markmyassignment has been
written. This makes it possible to get automatic feedback on specified assignments, using any computer
(altough internet access is required).
To install markmyassignment the package devtools is needed. To install devtools and markmyassign-
ment just run the following code:

> install.packages("devtools")
> devtools::install_github("MansMeg/markmyassignment")

To get automatic feedback on your assignment you need an assignment path from your teacher. To
set the assignment just run the following code

> library(markmyassignment)
> set_assignment("[assignment path]")

where [assignment path] äis the path given to you by your teacher.
To see which tasks that are included in the lab you can use the function show tasks() in the following
way:

> show_tasks()

To get automatic feedback you use the function mark my assignment(). To get feedback on all as-
signments just run the function.

> mark_my_assignment()

Remember that the functions need to be in the global environment in R.

You can also get feedback on specific tasks in the following way:

> mark_my_assignment(tasks="foo")
> mark_my_assignment(tasks=c("foo", "bar"))

It is also possible to correct an .R file in the following way:

> mark_my_assignment(mark_file = "[my search path to file]")

where [my search path to file] is the search path to the file to get feedback on.
Note! When an .R-file is checked, the global environment needs to be empty. Use rm(list=ls()) to
clean the global environment.

3
LINKÖPING UNIVERSITY
STIMA
Department of Computer and Information Science Advanced R Programming

Chapter 1

Lab assignments

To use markmyassignment run the following code:

library(markmyassignment)
lab_path <-
"https://raw.githubusercontent.com/STIMALiU/AdvRCourse/master/Labs/Tests/lab2.yml"
set_assignment(lab_path)

Assignment set:
Lab2: Advanced R programming, computer lab 2
The assignment contain the following (10) tasks:
- sheldon game
- my moving median
- for mult table
- find cumsum
- while mult table
- repeat find cumsum
- repeat my moving median
- in environment
- cov
- moment

1.1 Conditional statements

1.1.1 sheldon game(player1, player2)
In the series The big bang theory the character Sheldon comes up with a new version the game Rock-
paper-scissors, called Rock-paper-scissors-lizard-spock. The detailed rules of the game can be found here
(http://bigbangtheory.wikia.com/wiki/Rock Paper Scissors Lizard Spock).
Create a function you call sheldon game() that should take two arguments, player1 and player2.
Each player should choose one of the choices rock, paper, scissors, lizard or spock. If another choice
is made, the function should be stopped (with the stop() or stopifnot() function).
The function should return either ‘‘Player 1 wins!’’, ‘‘Player 2 wins!’’ or ‘‘Draw!’’.
Tip! This can be solved by using the cyclical structure of the game and the modulus operator.

> sheldon_game("lizard", "spock")

[1] "Player 1 wins!"

> sheldon_game("rock", "paper")

[1] "Player 2 wins!"

1.2 for loops
1.2.1 my moving median()
Create a function that will create a rolling median from a given numeric vector. Call the function
my moving median(). The functions should accept a given numeric vector x, and a numeric scalar n.
The function should check if the arguments is a numeric vector (x) and a numeric scalar (n) and if
not, it should stop the function (with the stop() or stopifnot() function). Otherwise the function
should return the following vector:

yt = median(xt , · · · , xt+n )

where yt is the tth elementet. It should also be possible to send argument to the median() function (as
na.rm). You should use a for loop in this assignment.

my_moving_median(x = 1:10, n=2)

[1] 2 3 4 5 6 7 8 9

my_moving_median(x = 5:15, n=4)

[1] 7 8 9 10 11 12 13

my_moving_median(x = c(5,1,2,NA,2,5,6,8,9,9), n=2)

[1] 2 NA NA NA 5 6 8 9

my_moving_median(x = c(5,1,2,NA,2,5,6,8,9,9), n=2, na.rm=TRUE)

[1] 2.0 1.5 2.0 3.5 5.0 6.0 8.0 9.0

1.2.2 for mult table()

Write a function called for mult table() with arguments from and to. Both arguments can be assumed
to be positive integers. The function should return a multiplication table as a matrix from the argument
from to the argument to. The column names and row names should contain the factors specified by the
arguments.
The function should stop (using stop()) if the arguments is not numeric scalars. You should use a
for loop in this assignment.
See an example below.

for_mult_table(from = 1, to = 5)

1 2 3 4 5
1 1 2 3 4 5
2 2 4 6 8 10
3 3 6 9 12 15
4 4 8 12 16 20
5 5 10 15 20 25

for_mult_table(from = 10, to = 12)

10 11 12
10 100 110 120
11 110 121 132
12 120 132 144

1.2.3 * cor matrix()

(NOTE: This task is Optional!)

5
 Create a function called cor matrix(). It should take a numeric data.frame with numerical variables
of an arbitrary size as an argument X and calculate the correlation matrix for the given variables. If
another object than an data.frame is used as argument, the function should return an error.
In this exercise you are not allowed to use the functions var(), sd() or cor() in your function. You
should use a for loop in this assignment.
See an example below.

data(iris)
cor_matrix(iris[,1:4])

[,1] [,2] [,3] [,4]

[1,] 1.00000 -0.11757 0.87175 0.81794
[2,] -0.11757 1.00000 -0.42844 -0.36613
[3,] 0.87175 -0.42844 1.00000 0.96287
[4,] 0.81794 -0.36613 0.96287 1.00000

data(faithful)
cor_matrix(faithful)

[,1] [,2]
[1,] 1.00000 0.90081
[2,] 0.90081 1.00000

1.3 while loops

1.3.1 find cumsum()
Create a functions that takes two arguments x and find sum. The function should calculate the cumula-
tive sum of the numeric vector x and stop when the cumulative sume is greater than find sum or if the
vector x is traversed and return this value. The function should assert that x is a numeric vector and is
a numeric scalar (with stop() or stopifnot()).
It is not allowed to usecumsum, a for loop in this assignment. You should use a while loop.

find_cumsum(x=1:100, find_sum=500)

[1] 528

find_cumsum(x=1:10, find_sum=1000)

[1] 55

1.3.2 while mult table()

Create a functions that calulates a given mutiplication table similar to the exercise 1.2.2 but now you
should use a (nested) while loop instead of a for-loop. The result should be identical in all other respects.

while_mult_table(from = 3, to = 5)

3 4 5
3 9 12 15
4 12 16 20
5 15 20 25

while_mult_table(from = 7, to = 12)

7 8 9 10 11 12
7 49 56 63 70 77 84
8 56 64 72 80 88 96

6
9 63 72 81 90 99 108
10 70 80 90 100 110 120
11 77 88 99 110 121 132
12 84 96 108 120 132 144

1.3.3 * trial division factorization()

(NOTE: This task is Optional!)
Write a function that can factorize any integer number x into its prime components. The function
should return a vector of all the√prime factors as a numeric vector. The idea is to simply test all integers
(or prime numbers) from 2 to x. For more detailed information on how to factorize primes using the
trial division algorithm can be found here (https://en.wikipedia.org/wiki/Trial division).

trial_division_factorization(x = 2^3 * 13 * 17 * 31)

[1] 2 2 2 13 17 31

trial_division_factorization(x = 47 * 91 * 97)

[1] 7 13 47 97

1.4 repeat and loop controls

1.4.1 repeat find cumsum()
Implement a function similar to find cumsum() in 1.3.1 but using repeat{} instead of a while loop. It
is not allowed to use cumsum in this exercise.

repeat_find_cumsum(x=1:100, find_sum=500)

[1] 528

repeat_find_cumsum(x=1:10, find_sum=1000)

[1] 55

1.4.2 repeat my moving median()

Implement a function similar to my moving median() in 1.2.1 but using repeat{} instead of a while
loop.

repeat_my_moving_median(x = 1:10, n=2)

[1] 2 3 4 5 6 7 8 9

repeat_my_moving_median(x = 5:15, n=4)

[1] 7 8 9 10 11 12 13

repeat_my_moving_median(x = c(5,1,2,NA,2,5,6,8,9,9), n=2)

[1] 2 NA NA NA 5 6 8 9

7
1.5 Environment
1.5.1 in environment()
Create a function called in environment() that returns the contents of an environment as a text vector.
The function should take the argument env that takes an environment name as a text element.

env <- search()[length(search())]

env

[1] "package:base"

funs <- in_environment(env)

funs[1:5]

[1] "!" "!.hexmode" "!.octmode" "!=" "$"

1.5.2 * where()
(NOTE: This task is Optional!)
This function is similar to the where() function described in Advanced R. But unlike the implemen-
tation in the book you are not allowed to use parent.env() to traverse the search path. The function
should take a name and return the name of the environment where the object is found by following the
search path.
If the function is not found, the function should return the text element ‘‘[fun] not found!’’.
Assert that fun is a character vector of length 1.

where(fun = "sd")

[1] "package:stats"

where(fun = "read.table")

[1] "package:utils"

where(fun = "non_existant_function")

[1] "non_existant_function not found!"

1.6 Functionals
1.6.1 cov()
This exercise is taken from the Advanced R book. Use lapply() and an anonymous function to find
the coefficient of variation (the standard deviation divided by the mean) for all columns in a given
data.frame. The function should return a vector of the coefficients of variation.
Assert that X is a data.frame.

data(iris)
cov(X = iris[1:4])

Sepal.Length Sepal.Width Petal.Length Petal.Width

0.14171 0.14256 0.46974 0.63555

cov(X = iris[3:4])

Petal.Length Petal.Width
0.46974 0.63555

8
1.7 Closures
1.7.1 moment()
This exercise has been taken from the Advanced R book. Create a function that creates a function
for the ith central moment for a given variable. Definitions of the central moments can be found here
(https://en.wikipedia.org/wiki/Central moment).

m1 <- moment(i=1)
m2 <- moment(i=2)
m1(1:100)

[1] 0

m2(1:100)

[1] 833.25

1.7.2 * mcmc counter factory()

(NOTE: This task is Optional!)
In Markov chain Monte Carlo methods we often want to use counters for the sampling algorithms.
Create a counter factory that creats a closure with a counter. The factory should take two arguments,
thin and burnin. Each counter should contain three objects:
• iteration (numeric) scalar

• store sample (boolean) scalar

• samples (numeric) scalar
Assume that the counter is called in each iteration so in each call of the counter (without argument)
should bump up the iterations object.
The store sample boolean indicates whether the current iterations should be stored. Iterations after
the number of burnin iterations should be stored every thin iteration. So if we run 20 iterations with
burnin = 10 and thin = 3 we would store iteration {13, 16, 19}. The samples scalar should keep track of
the number of stored iterations (three as the example). In each iteration the values should be returned
as a list.
Assert that burnin is a nonnegative value and that thin is a positive value.
Below is an example.

mcmccnt <- mcmc_counter_factory(burnin = 3, thin = 2)

mcmccnt()

[[1]]
[1] 1

[[2]]
[1] FALSE

[[3]]
[1] 0

mcmccnt()

[[1]]
[1] 2

[[2]]
[1] FALSE

9
[[3]]
[1] 0

mcmccnt()

[[1]]
[1] 3

[[2]]
[1] FALSE

[[3]]
[1] 0

mcmccnt()

[[1]]
[1] 4

[[2]]
[1] FALSE

[[3]]
[1] 0

mcmccnt()

[[1]]
[1] 5

[[2]]
[1] TRUE

[[3]]
[1] 1

mcmccnt()

[[1]]
[1] 6

[[2]]
[1] FALSE

[[3]]
[1] 1

10