Special topics in finance lecture 2

THE R LANGUAGE FOR
STATISTICAL COMPUTING
Lecture 2
Dr. Syed
Muhammad Ali
Tirmizi
Chapter No. 2
1

2
INTRODUCTION
 R language has its roots at AT&T Bell
Laboratories during the 1970s and 1980s in the
S language project (Becker, Chambers, and
Wilks, 1988).
 People think that the S language would not have
been designed in the way it was if it had been
designed by computer scientists (Morandat, Hill,
Osvald, and Vitek, 2012).

3
INTRODUCTION
 R is an open source variant of S developed at the
University of Auckland by Ross Ihaka and Robert
Gentleman, first appearing in 1993 (Ihaka, 1998).
 Clearly the recent popularity of R, fueled by its
open source availability and the need for statistical
and analytical computing tools, shows that the
benefits of R far outweigh the negatives.
 Overall, R is based upon the vector as a first class
item in the language.

4
INTRODUCTION
 R shares this attribute with LISP, Scheme,
Python, and Matlab.
 This and the prevalence of over 4,000 publicly
available packages are two of the many
strengths of R.
 In this course, we will focus on R packages that
revolve around financial analytics.

5
GETTING STARTED WITH R
 One of the great things about R is how easy it is to install.
 In your browser, head to the web site for the Comprehensive R
Archive Network (CRAN), http://cran.r-project.org
 This website will help you to download and install R software for
windows.
 Just as a basic test, we can create a vector of prices and plot it with
this block of code:
x = c(1.3,1.2,1.3,NA,1.4,1.5)
plot(x,ylab="EUR prices")
is.na(x)
[1] FALSE FALSE FALSE TRUE FALSE FALSE

6
 The c() operator creates a vector of elements.
 This is the basic vector operator in R.
 Note the “not available” (NA) element
appearing as the fourth item of the vector.
 R’s ability to handle NAs, infinite values (Inf),
and not a number (NaN) is one of its many
strengths.

7
 A later development is the R Studio GUI available
from the web site www.rstudio.com.
 R Studio is a commercially developed GUI allowing
management of plot windows, variable contents,
and better debugging than the basic R interpreter.
 R Studio is a second-generation R user interface
with integrated code, execution, variable inspection,
and plotting windows, and expression completion.

9
 Any time a library statement is encountered, R
will check that the package is available.
 If not, it must be downloaded.
 As an example, to download the ggplot2
package, use the following command:
update.packages()
install.packages("ggplot2",dependencies=TRUE)
library(ggplot2)

10
LANGUAGE FEATURES: FUNCTIONS,
ASSIGNMENT, ARGUMENTS, AND TYPES
 For many use cases, R provides a computational
statistics platform.
 Mathematical functions are readily available. The
basic log() function provides a natural logarithm.
Of course, executing log() on a vector, x, results
in a vector of natural logarithms, y.

11
 The computation of log() on NA is NA as expected.
> #Filter prices:
> x[x > 1.3]
[1] NA 1.4 1.5
> #Keeps the NA intact:
> y <- diff(log(x))
> round(y,3)
[1] -0.080 0.080 NA NA 0.069

12
 In R, not only vectors but also functions are first-
class objects. It shares this attribute with the
functional languages LISP and Scheme.
 Assigning a function to a variable is the usual
way to define it. If g is assigned to the function
definition, then g(4) will evaluate it and g,
without parentheses, will return its definition.

13


14
 R has four assignment operators.
 The most basic operator is “<-”. This is the one
we use in the first assignment in the code block
below.
 R’s functional nature is so strong that even this
can be replaced by the function call assign(“x”,1).

15
 Over time, because people were used to other
languages that use “=” instead, and even though
“=” was used to assign parameter values in
function calls (g(x, y = 7), for example), it was
also made available for assignment in R.
 So using “<-” or “=” is now really a matter of
preference.

16
> x <- 1
> assign("x",2)
> x
[1] 2
> x = 3
> x
[1] 3
> f <-function(x)
+ {
+ x = 4
+ x
+ }
> f(x)
[1] 4
> x
[1] 3
RUN THESE
CODES

17
 R’s fourth assignment operator, with two “<”s, is
known as the “super-assignment” operator.
 Executing it will look outside the current frame
for x, which is global to the function f , and
assign to that x.
 If there is no x in the global environment, it will
create one and assign the value to it.

18
> #The fourth type is "<<-"
> x = 3
> x
[1] 3
> f <-function(x)
+ {
+ x <<- 4
+ x
+ }
> f(x)
[1] 4
> x
[1] 4
> typeof(f)
[1] "closure"
> typeof(x)
[1] "double"
RUN THESE
CODES

19
 R is dynamically typed so that variables do not have types.
 Instead, values have types.
 So we can see that the type of a variable is determined by
the type of its current value.
 The function typeof() can be used to return the type of the
value assigned to a variable.
 We can see its use in the output block above: typeof(f) is a
“closure” for a function while typeof(x) is a “double.”

20
LANGUAGE FEATURES:
BINDING AND ARRAYS
 Binding scalars or vectors together is one way to return
aggregate results from functions.
 For binding, cbind() binds items into two columns and
rbind() binds items into two rows.
 If the two items are scalars then the two operations are
equivalent. rep() is a very common function to create a
vector of repeated items.
 For example, rep(4, 5) == c(4, 4, 4, 4, 4) is TRUE and
states to repeat 4 five times.

21
LANGUAGE FEATURES:
BINDING AND ARRAYS
> #Create two column matrix:
> A = cbind(rep(x,length(y)),y)
> A
y
[1,] 4 -0.08004271
[2,] 4 0.08004271
[3,] 4 NA
[4,] 4 NA
[5,] 4 0.06899287
> B = rbind(rep(x,length(y)),y)
> B
[,1] [,2] [,3] [,4] [,5]
4.00000000 4.00000000 4 4 4.00000000
y -0.08004271 0.08004271 NA NA 0.06899287
> t(A) == B
[,1] [,2] [,3] [,4] [,5]
TRUE TRUE TRUE TRUE TRUE
y TRUE TRUE NA NA TRUE
> sum(t(A) == B)
[1] NA
RUN THESE
CODES

22
ERROR HANDLING
 Error handling is an important part of data science, in
order to keep erroneous data from making its way into the
variables and to keep the dataset as clean as possible.
 When calling certain packages, it is common to have
errors returned. R has a tryCatch() feature which is
implemented as a function.
 Example, we use division by zero as the type of error
handled.

23
ERROR HANDLING
#Exception handling:
fh <- 0
tryCatch({
#main block
fh <<- file("file1.txt", open="r")
}, warning = function(w) {
#warning-handler-code
print(w)
fh <<- NA
}, error = function(e) {
#error-handler-code
print(e)
fh <<- NA
}, finally = {
#cleanup-code
})
if(!is.na(fh)) readLines(fh)
RUN THESE
CODES

24
NUMERIC, STATISTICAL,
AND CHARACTER FUNCTIONS
 To set the mode of calculations, options(digits = n) sets
the number of digits to round to in calculations.
#Setting precision:
options(digits=10)
pi = 3.1415926535897932384626
pi
 [1] 3.141592654

25
 Distributions of random variates are available readily
within the language with runif() for the Uniform
Distribution, rnorm() for the Normal Distribution, and
rbinom() for the Binomial Distribution, to name just
three.
 Histograms and density plots are also native to the
language with hist() and density().

26
plot(density(rbinom(50,50,1/2))
options(digits=6)
set.seed(99)
sample(10, replace = TRUE)
[1] 6 2 7 10 6 10 7 3 4 2

27
DATA FRAMES AND INPUT–OUTPUT
 Data frames are one of R’s more unusual and handy
features.
 A data frame is a sequence of rows where the columns are
heterogeneously typed. A common way of loading data
frames is from the Excel .csv files.
#Input-ouput:
write.csv(d,file="d.txt",row.names=FALSE)
e <- read.csv("d.txt",header=TRUE)

28
LISTS
 Lists are ordered aggregates like vectors, which are
constructed by c(. . .).
 However, lists differ from the basic vectors, in that they
are recursively formed, using list(. . .).
#Lists:
c(1,c(1,2),3,"A",c(4,5))
[1] "1" "1" "2" "3" "A" "4" "5"
list(1,c(1,2),3,"A",list(4,5))

Special topics in finance lecture 2

More Related Content

What's hot (20)

Similar to Special topics in finance lecture 2 (20)

More from Dr. Muhammad Ali Tirmizi., Ph.D. (19)

Recently uploaded (20)

Special topics in finance lecture 2