Welcome to the R intro Workshop Before we begin, please download the “SwissNotes.csv” and “cardiac.txt” files from the ISCC website, under the Introduction.

Slides:



Advertisements
Similar presentations
Summary Statistics/Simple Graphs in SAS/EXCEL/JMP.
Advertisements

Introduction to R Brody Sandel. Topics Approaching your analysis Basic structure of R Basic programming Plotting Spatial data.
R for Macroecology Aarhus University, Spring 2011.
Welcome to the R intro Workshop Before we begin, please download the “SwissNotes.csv” and “cardiac.txt” files from the ISCC website, under the R workshop.
Statistical Methods Lynne Stokes Department of Statistical Science Lecture 7: Introduction to SAS Programming Language.
Introduction to Matlab Workshop Matthew Johnson, Economics October 17, /13/20151.
R tutorial g/methods2.2010/R-intro.pdf.
 Statistics package  Graphics package  Programming language  Can be used to share/reproduce analyses  Many new packages being created - can be downloaded.
Basics of Using R Xiao He 1. AGENDA 1.What is R? 2.Basic operations 3.Different types of data objects 4.Importing data 5.Basic data manipulation 2.
Introduction to GTECH 201 Session 13. What is R? Statistics package A GNU project based on the S language Statistical environment Graphics package Programming.
A Simple Guide to Using SPSS© for Windows
MATLAB TUTORIAL Dmitry Drutskoy Some material borrowed from the departmental MATLAB info session by Philippe Rigollet Kevin Wayne.
Introduction to MATLAB MECH 300H Spring Starting of MATLAB.
How to Use the R Programming Language for Statistical Analyses Part I: An Introduction to R Jennifer Urbano Blackford, Ph.D. Department of Psychiatry Kennedy.
SPSS Statistical Package for the Social Sciences is a statistical analysis and data management software package. SPSS can take data from almost any type.
Introduction to SPSS Short Courses Last created (Feb, 2008) Kentaka Aruga.
LISA Short Course Series Basics of R Lin Zhang Feb. 16, 2015 LISA: Basics of RFeb. 16, 2015.
Introduction to MATLAB ENGR 1187 MATLAB 1. Programming In The Real World Programming is a powerful tool for solving problems in every day industry settings.
Basic R Programming for Life Science Undergraduate Students Introductory Workshop (Session 1) 1.
Chapter 5 Review: Plotting Introduction to MATLAB 7 Engineering 161.
Quantitative Research in Education Sohee Kang Ph.D., lecturer Math and Statistics Learning Centre.
ALEXANDER C. LOPILATO R: Because the names of other stat programs don’t make sense so why should this one?
732A44 Programming in R.  Self-studies of the course book  2 Lectures (1 in the beginning, 1 in the end)  Labs (computer). Compulsory submission of.
Hands-on Introduction to R. Outline R : A powerful Platform for Statistical Analysis Why bother learning R ? Data, data, data, I cannot make bricks without.
Data, graphics, and programming in R 28.1, 30.1, Daily:10:00-12:45 & 13:45-16:30 EXCEPT WED 4 th 9:00-11:45 & 12:45-15:30 Teacher: Anna Kuparinen.
1 Experimental Statistics - week 4 Chapter 8: 1-factor ANOVA models Using SAS.
Introduction to R Part 2. Working Directory The working directory is where you are currently saving data in R. What is the current working directory?
Arko Barman with modification by C.F. Eick COSC 4335 Data Mining Spring 2015.
1 Lab of COMP 406 Teaching Assistant: Pei-Yuan Zhou Contact: Lab 1: 12 Sep., 2014 Introduction of Matlab (I)
ECE 1304 Introduction to Electrical and Computer Engineering Section 1.1 Introduction to MATLAB.
Computational Methods of Scientific Programming Lecturers Thomas A Herring, Room A, Chris Hill, Room ,
Input, Output, and Processing
Piotr Wolski Introduction to R. Topics What is R? Sample session How to install R? Minimum you have to know to work in R Data objects in R and how to.
I❤RI❤R Kin Wong (Sam) Game Plan Intro R Import SPSS file Descriptive Statistics Inferential Statistics GraphsQ&A.
Using the R software R is an open source comprehensive statistical package, more and more used around the world. R project web site:
Installing R CRAN: –(R homepage: –Windows 95 and later  Base –rw2001.exe.
Hands-on Introduction to R. We live in oceans of data. Computers are essential to record and help analyse it. Competent scientists speak C/C++, Java,
R Programming Yang, Yufei. Normal distribution.
Introduction to SPSS. Object of the class About the windows in SPSS The basics of managing data files The basic analysis in SPSS.
Introduction to Programming in R Department of Statistical Sciences and Operations Research Computation Seminar Series Speaker: Edward Boone
Chapter 1 – Matlab Overview EGR1302. Desktop Command window Current Directory window Command History window Tabs to toggle between Current Directory &
STAT 251 Lab 1. Outline Lab Accounts Introduction to R.
Introduction to R Introductions What is R? RStudio Layout Summary Statistics Your First R Graph 17 September 2014 Sherubtse Training.
Basics of Biostatistics for Health Research Session 1 – February 7 th, 2013 Dr. Scott Patten, Professor of Epidemiology Department of Community Health.
Lecture 26: Reusable Methods: Enviable Sloth. Creating Function M-files User defined functions are stored as M- files To use them, they must be in the.
Introduction to R Carol Bult The Jackson Laboratory Functional Genomics (BMB550) Spring 2011.
Chapter 6: Analyzing and Interpreting Quantitative Data
Mr. Magdi Morsi Statistician Department of Research and Studies, MOH
STAT 534: Statistical Computing Hari Narayanan
Overview Excel is a spreadsheet, a grid made from columns and rows. It is a software program that can make number manipulation easy and somewhat painless.
INTRODUCTION TO MATLAB Dr. Hugh Blanton ENTC 4347.
Digital Image Processing Introduction to MATLAB. Background on MATLAB (Definition) MATLAB is a high-performance language for technical computing. The.
R objects  All R entities exist as objects  They can all be operated on as data  We will cover:  Vectors  Factors  Lists  Data frames  Tables 
Math 252: Math Modeling Eli Goldwyn Introduction to MATLAB.
Hands-on Introduction to R. We live in oceans of data. Computers are essential to record and help analyse it. Competent scientists speak C/C++, Java,
Lecture 11 Introduction to R and Accessing USGS Data from Web Services Jeffery S. Horsburgh Hydroinformatics Fall 2013 This work was funded by National.
Working with data in R 2 Fish 552: Lecture 3. Recommended Reading An Introduction to R (R Development Core Team) –
EMPA Statistical Analysis
Programming in R Intro, data and programming structures
Introduction to R Samal Dharmarathna.
By Dr. Madhukar H. Dalvi Nagindas Khandwala college
DEPARTMENT OF COMPUTER SCIENCE
MATLAB DENC 2533 ECADD LAB 9.
Use of Mathematics using Technology (Maltlab)
Communication and Coding Theory Lab(CS491)
MIS2502: Data Analytics Introduction to R and RStudio
R Course 1st Lecture.
Stat 251 (2009, Summer) Lab 2 TA: Yu, Chi Wai.
Data analysis with R and the tidyverse
R tutorial
Presentation transcript:

Welcome to the R intro Workshop Before we begin, please download the “SwissNotes.csv” and “cardiac.txt” files from the ISCC website, under the Introduction to R workshop to the desktop of your computer. 1

Introduction to R Workshop in Methods from the Indiana Statistical Consulting Center Thomas A. Jackson October 11,

Overview The R Project for Statistical Computing “R is a language and environment for statistical computing and graphics. It is a GNU project which is similar to the S language and environment which was developed at Bell Laboratories (formerly AT&T, now Lucent Technologies) by John Chambers and Colleagues. R can be considered as a different implementation of S. There are some important differences, but much code written for S runs unaltered under R.” - Description from CRAN Website 3

Benefits R … is free is interactive: we can type something in and work with it ▫How we analyze data can be broken into small steps is interpretative: we give it commands and it translates them into mathematical procedures or data management steps can be used in a batch: nice because it is documented is a calculator: it is unlike other calculators though because you can create variables and objects 4

Let’s Get R Started How to open R → Start Menu → Programs → Departmentally Supported → Stat/Math → R 5

Graphical User Interface (GUI) Three Environments Command Window (aka Console) Script Window Plot Window 6

Command Window Basics To quit: type q() Save workspace image? Moves from memory to hard- drive Storing variable in memory, or = a<- 5 stores the number 5 in the object “a” pi -> b stores the number π= in “b” x = stores the result of the calculation (3) in “x” “=“ requires left-hand assignment Try not to overwrite reserved names such as t, c, and pi! 7

Command Window Basics Printing to output Calculations that are not stored print to output > [1] 8 Type name to view stored object > a [1] 5 Use print() > print(a) [1] 5 View objects in workspace objects() or ls() 8

Command Window Basics Clearing the console (command window) Mac: Edit → Clear Console Windows: Edit → Clear Console or Mac: Alt + Command + L Windows: Ctrl + L Removing variables from memory rm() or remove() > x <- 4 > rm(x) rm(list = ls()) remove all variables 9

Script Window Basics Saving syntax (code) Mac: File → New Windows: File → New Script Documenting code: # Comments out everything on line behind Running code from Script Window Mac: Apple + Enter Windows: F5 or Ctrl + r 10

Working Directory Obtaining working directory getwd() Mac: Misc → Get Working Directory Windows: File → Change dir... Changing working directory setwd() Mac: Misc → Change Working Directory Windows: File → Change dir... 11

Path Names Specify with forward slashes or double backslashes Enclose in single or double quotation marks Examples setwd(“C:/Users/jacksota/Desktop”) setwd(‘C:\\Users\\jacksota\\Desktop’) 12

Try it! #1 1)From the command window find your current working directory. Change the working directory to be the Desktop folder under your Username. 2)Save the commands for finding and changing the working directory to the desktop in a script file. Save the script to the desktop. 13

R Help Helpful commands If you know the function name: help() or ? > help(log) > ?exp If you do not know the function name: help.search() or ?? > help.search(“anova”) > ??regression 14

Documentation Elements of a documentation file Function{Package} Description Usage: What your code should look like, “=“ gives default Arguments: Inputs to the function Details Value: What the function will return See Also: Related functions Examples 15

Online Resources CRAN Website: R Seek: Quick-R tutorial: R Tutor: UCLA: R listservs Google Google tip: include “[R]” (instead of just “R”) with search topic to help filter out non-R websites 16

Additional Packages Over 4,900 listed on the CRAN website! Use with caution Initial download of R: base, graphics, stats, utils 1) Installing a package: Mac: Packages & Data → Package Installer Use Package Search to locate and press ‘Install Selected’ Windows: Packages → Install Packages Locate desired package and press ‘OK’ install.packages(“MASS”) 2) Using an installed package: You MUST call it into active memory with library() > library(MASS) 17

Try it! #2 1) Using help() or ?, open the documentation for plot table eigen 2) Locate and install the psych package 18

Data Structures R has several basic types (or “classes”) of data: Numeric - Numbers Character – Strings (letters, words, etc.) Logical – TRUE or FALSE Vector Matrix Array Data Frame List NOTE: There are other classes, but these are most common. Understanding differences will save you some headache. 19

Data Structures Find class of data Unknown class: class() Check particular class: is.“classname”() > a <- 5 > class(a) [1] “numeric” > is.character(a) [1] FALSE Change class: as.classname() > as.character(a) [1] “5” 20

Vectors Combine items into vector: c() > c(1,2,3,4,5,6) [1] Repeat number of sequence of numbers: rep() > rep(1,5) [1] > rep (c(2,5,7), times = 3) [1]

Vectors Sequence generation: seq() > seq(1,5) [1] > seq(1,5, by =.5) [1] Try 1:10 or 10:1 22

Matrices Create matrix: matrix() 6 x 1 matrix: matrix(1:6, ncol = 1) 2 x 3 matrix: matrix(1:6, nrow =2, ncol =3) 2 x 3 matrix filling across rows first: matrix(1:6, nrow = 2, ncol = 3, byrow = TRUE) Create matrix of more than two dimensions (array): array() 23

Lists Create a list: list() Holds vectors, matrices, arrays, etc. of varying lengths Objects in the list can be named or unnamed > list(matrix(0, 2, 2), y = rep(c(“A”, “B”), each = 2)) [[1]] [,1] [,2] [1,]00 [2,]00 $y [1] “A” “A” “B” “B” Data Frame: specialized list that holds variables of same length 24

Data Frames Create a data frame: data.frame() Like a matrix, holds specified number of rows and columns > x <- 1:4 > y <- rep(c(“A”, ”B”), each = 2) > data.frame(x,y) x y 1 1 A 2 2 A 3 3 B 4 4 B Unnamed variables get assigned names > data.frame(1:2, c(“A”, “B”)) X1.2 c..A….B A 2 2 B 25

Try It! #3 Enter the following data into R: CHI FRA GER ) Try to enter the data above as a data frame 2) Can you enter it as a matrix? Would the data be different in a matrix, rather than a dataframe? 26

Basic Operations Arithmetic: +, -, *, / Order of operations: () Exponentiaition: ^, exp() Other: log(), sqrt Evaluate standard Normal density curve, at x = 3 > x <- 3 > 1/sqrt(2*pi)*exp(-(x^2)/2) [1]

Vectorization R is great at vectorizing operations Feed a matrix or vector into an expression Receive an object of similar dimension as output For example, evaluate at x = 0,1,2,3 > x <- c(0,1,2,3) > 1/sqrt(2*pi)*exp(-(x^2)/2) [1]

Logical Operations Compare: ==, >, =, <=, != > a <- c(1,1,2,4,3,1) > a == 2 [1] FALSE FALSE TRUE FALSE FALSE FALSE And: & or && Or: | or || Find location of TRUEs: which() > which(a == 1) [1]

Subsetting > a <- 1:5 > b <- matrix(1:12,nrow = 3) Use Square brackets [] Pick range of elements: a[1:3] Pick particular elements: a[c(1,3,5)] Do not include elements: a[-c(1,4)] 30

Subsetting (cont.) Use commas in more than on dimension (matrices & data frames) Pick particular elements: B[1:2,2:4] Give all rows and specified columns: B[,1:2] Give all columns and specified rows: B[1:2,] Note: B[2] coerces into a vector then gives specified element 31

Reading External Data Files SwissNotes.csv Data set Complied by Bernard Flury Contains measurements on 200 Swiss Bank Notes 100 genuine and 100 counterfeit notes 32

Reading External Data Files (cont.) Most general function: read.table() read.table(file,header=FALSE,sep = “”,…) Creates a data frame File name must be in quotes, single or double File name is case sensitive Include file name extension if data not in working directory > read.table(“C:\\Users\\jacksota\\Desktop\\SwissNotes.csv”, T,“,”) > read.table(“SwissNotes.csv”,T,“,”), if we have set the working dir. correctly Don’t know the file extension? Try: file.choose() > read.table(file.choose(), header = TRUE, sep = ”,”) sep defines the separator, e.g. “,” or “\t” or “” header indicates variable names should be read from first row 33

Reading External Data Files For comma delimited files: read.csv() For tab delimited files: read.delim() For Minitab, SPSS, SAS, STATA, etc. data: foreign package Contains functions to read variety of file formats Functions operate like read.data() Contains functions for writing data into these file formats 34

Data Frame Hints Identify variable names in data frame: names() > data1 <- read.table(“SwissNotes.csv”, sep=“,”, header =TRUE) > names(data1) [1] “Length” “LeftHeight” “RightHeight” “LowerInner.Frame” [5] “UpperInner.Frame” “Diagonal” “Type” Assign name to data frame variables > names(data1) <- c(“Length”, “LeftHeight”, “RightHeight”, “LowerInner..Frame”, “UpperInner.Frame”, “Diagonal”, “Type”) Note: names are strings and MUST be contained in quotes 35

Data Frame Hints (cont.) Create objects out of each data frame variable: attach() In the Swiss Note data, to refer to Type as its own object > attach(data1) > Type [1] GenuineGenuineGenuine …. 36

Data Frame Hints (cont.) Remove attached objects from workspace: detach() > detach(data1) > Type Error: object “Type” not found Note: Type is still part of original data frame, but is no longer a separate object. 37

Try It! #4 Caridac.txt includes measurements taken to predict cardiac events, such as heart attacks. You should have already downloaded it and saved to your desktop Tab delimited data set (sep=“\t”) with header 1)Read cardiac.txt into R 2)Select data for male participants only 3)For male participants, select data for basal heart rate (bhr) and peak heart rate (pkhr) only 38

plot() function plot() is the primary plotting function Calling plot will open a new plotting window Documentation: ?plot For complete list of graphical parameters to manipulate: ?par 39

plot() function Let’s visualize the SwissNotes.csv data. After loading the data into R, attach the data frame using attach(data). Let’s try a scatter plot of LeftHeight by RightHeight. >plot(LeftHeight, RightHeight) 40

plot() function Change symbols: Option pch=. See ?par for details. >plot(LeftHeight,RightHeight,pch=2) 41

plot() Function Change symbol color: Option col= Specify by number or by name: col=2 or col=“red” Hint: Type palette() to see colors associated with number Type colors() to see all possible colors > plot(LeftHeight, RightHeight, col=“red”) 42

What types of points can we get? 43

plot() Function Change plot type: Option type = “p” for points “l” for lines “b” for both “c” for lines part alone of “b” “o” for both overplotted “h” for histogram like (or high-density) vertical lines “s” for stair steps “S” for other steps, see Details below “n” for no plotting 44

45

Plot() Function Points with lines…works better on sorted list of points >plot(LeftHeight,RightHeight,type=“o”) 46

Scatterplots for Multiple Groups Use plot() with points() to plot different groups in same plot Genuine notes vs. Counterfeit notes >plot(LeftHeight[Type==“Genuine”],Rightheight[Type==“Genuine”], col=“red”) >points(LeftHeight[Type==“Counterfeit”],RightHeight[Type==“Counterfeit”],col=“blue”) 47

Axis Labels and Plot Titles The plot() command call has options to Specify x-axis label: xlab = “X Label” Specify y-axis label: ylab = “Y Label” Specify plot title: main = “Main Title” Specify subtitle: sub = “Subtitle” 48

Axis Labels and Plot Titles >plot(LeftHeight[Type==”Genuine”],RightHeight[Type==“Genuine”], col=“red”,main=“Plot of Bank Note Heights”,sub=“Measurements are in mm”,xlab=“Height of Left Side”,ylab=“Height of Right Side”) >points(LeftHeight[Type==“Counterfeit”], RightHeight[Type=“Counterfeit”],col=“blue”) 49

Legends  legend(“topleft”,c(“Genuine Notes”, ”Counterfeit Notes”),pch=c(21,21),col=c(“red”,”blue”)) 50

Try It! Using the cardiac data set, try the following exercises 1)Create a scatterplot of age vs. peak heart rate (pkhr). Be sure to label your axes and title the plot. 2)Using information available on the points() documentation page, adjust your scatterplot to have symbols that are filled blue squares. 3)In your age versus peak heart rate scatterplot, plot the males and females with different symbols and different colors. Add a legend to your plot. 51

Adding Lines To add straight lines to plot: abline() abline() refers to standard equation for a line: y = bx + a Horizontal line: abline(h= ) Vertical Line: abline(v= ) Otherwise: abline(a=, b= ) or abline(coef=c(a,b)) 52

Adding Lines > abline(coef=c( ,0.8319)) 53

Histograms Histograms are another popular plotting option. > hist(Length) 54

pairs() Function Using the SwissNote Data > pairs(swiss) 55

Boxplots To create boxplots: boxplot() Specify one or more variables to plot. > boxplot(swiss$Length) > boxplot(swiss[,2:3]) 56

Boxplots Use a formula specification for side-by-side boxplots. Note: boxplot() has many options, e.g. notches. See ?boxplot. > boxplot(Length~Type,notch=TRUE,data=swiss) 57

Try It! Using the cardiac data set, try the following exercises 1)Create histograms for basal blood pressure (basebp) and systolic blood pressure (sbp). 2)Try creating boxplots of the same variables. Can you do separate plots for Males and Females? 58

Mean or Average Mean() > mean(swiss[,”Length”]) > mean(swiss) rowMeans() > rowMeans(swiss[,1:6]) colMeans > colMeans(swiss[,7]) 59

Variability Variance: var() > var(swiss[,”Length”]) > var(swiss) Covariance() > cov(swiss) Correlation() > cor(swiss[,1:6]) Standard deviation > sd(swiss$length) 60

Five-number Summary >summary(swiss[1:3]) Length LeftHeight RightHeight Min. :213.8 Min. :129.0 Min. : st Qu.: st Qu.: st Qu.:129.7 Median :214.9 Median :130.2 Median :130.0 Mean :214.9 Mean :130.1 Mean : rd Qu.: rd Qu.: rd Qu.:130.2 Max. :216.3 Max. :131.0 Max. :

Creating Tables table() produces crosstabs of factors or categorical variables Using the cardiac data: > table(cardiac[,7:9]),, newMI = 0 chestpain gender 0 1 F 6 10 M 4 8,, newMI = 1 chestpain gender 0 1 F M

Try It! Using the cardiac data Find the descriptive statistics for the chestpain (chestpain) and dose (dose). Note that chestpain is categorical (0=no chest pain, 1=chest pain) and that does is continuous. 63

Univariate t-tests t.test() produces 1- and 2-sample (paired or independent) t- tests. 1-sample t-test > t.test(x,alternative=“two.sided”,mu=0,conf.level=0.95) 2 independent samples t-test > t.test(x,y,alternative=“two.sided”,mu=0,paired=FALSE, conf.level=0.95) paired t-test > t.test(x,y,alternative=“two.sided”,mu=0,paired=TRUE, var.equal=TRUE,conf.level=0.95) 64

2 Independent Samples t-test x: diagonal measurements for Genuine bank notes y: diagonal measurements for Counterfeit bank notes > x = swiss[Type==“Genuine”,”Diagonal”] > y = swiss[Type==“Counterfeit”,”Diagonal”] > t.test(x,y,alternative=“greater”,mu=0, paired=FALSE,var.equal=TRUE) 65

2 Independent Samples t-test > t.test(x,y,alternative=“greater”,mu=0, paired=FALSE,var.equal=TRUE) Two Sample t-test data: x and y T = , df = 198, p-value < 2.2e-16 alternative hypothesis: true difference in means is greater than 0 95 percent confidence interval: Inf sample estimates: mean of x mean of y

Try It! #8 Using the cardiac data Using the appropriate t-tests, test the following hypotheses H 0 : μ age = 72 vs. H 1 : μ age < 72 For peak heart rate (pkhr), H 0 : μ male = μ female vs. H 1 : μ male > μ female For peak heart rate, H 0 : μ no chest pain = μ chest pain vs. H 1 : μ no chest pain ≠ μ chest pain 67

Generating Random Numbers R contains functions for generating random numbers from many well-known distributions. Random number from standard normal distribution: > rnorm(1,mean=0,sd=1) [1] Vector of random numbers from uniform distribution: > runif(3, min=0, max=1) [1] To reproduce results: set.seed() 68

Function Basics if() statement > n = rnorm(1) > if(n < 0){ n = abs(n) } if() statement with else() > n = rnorm(1) >if (n < 0){ n = abs(n) } else{n = 0} 69

Function Basics for() loop > temp = rep(0,10) > for (i in 1:10){ temp[i] = i+1 } > temp [1]

Function Basics while() loop > n = 1 > while (n < 10 ){ n = n+1 } 71

Creating Functions test.function = function(input arguments){ commands to execute } 72

Creating Functions For example, let’s define a new function average to find the average of a set of numbers. average = function(x){ n = length(x) average = sum(x)/n print(average) } 73

Sourcing After writing a function in a script file, bring it into working memory using source(). Source(“pathname/test.function.R”) 74