🧠

Search

SearchSearch

R Programming

Apr 03, 2024, 3 min read

R Programming

Different from S? Sweave

01 Background

Resources

  • 📖 R for Data Science ⭐️⭐️⭐️⭐️⭐️
  • 📖 The Book of R
  • 📖 The Art of R Programming
  • 📹 Data Science: Foundations using R Specialization by Coursera ⭐️⭐️⭐️
  • 🌐 R for Beginners by CRAN
  • https://r-graph-gallery.com/

Limited in big data, too big to load in memory

02 Core Concepts

Key Terms

  • Facets
  • Geoms
  • Asethetic
  • Atomic: (1) character (2) number (real) (3) integer (4) complex (5) logical Graphs tutorial
  • Attributes: (1) names, dimnames (2) dimensions (matrices, arrays) (3) class (4) length (5) other user-defined attributes/metadata
  • Coercion Implicit / Explicit
  • Matrices
  • Factors Ordered / Unordered
  • Missing Values NaN is Na, but Na is not NaN
  • Data Frames
  • Reading Data
    • read.table and read.csv for tabular
    • source for reading R code files
  • Dput, Dget, Dumping
  • Subsetting
  • Partial Matching
  • https://www.storybench.org/getting-started-data-visualization-r-using-ggplot2/
  • cacher package
  • Plotting base, lattice, ggplot2, tidyverse
dir
?args  # helper
args() # determine the arguments of a

”...” argument used heavily in generic functions

Lexical Scoping

Complete Cases Times and Dates

  • as.Date for dates
  • as.POSIXct and as.POSIXlt for times
  • strptime can be used to coerce to the above

Loop Functions (to make looping more easy on the command line)

  • lapply look over a list and evaluate a function on each element
  • sapply same as lapply but try to simplify the result
  • apply apply function over the margins of array
  • tapply apply function over subsets of a vector
  • mapply multivariate version of lappy

Factor Variables

Debugging Tools - Basic Tools

  • traceback
  • debug
  • browser
  • trace
  • recover

str Function

  • compactly display the internal structure of an R object
  • an alternative to summary

Simulation

  • rnorm generate random Normal variates with a given mean and standard deviation
  • dnorm evaluate the Normal probability (with a given mean/SD) at a point (or vector of points)
  • pnorm evaluate the cumulative distribution function for a Normal distribution
  • rpois generate random Poisson variates with a given rate

R Profiler

  • system.time()
  • Rprof
  • summaryRprof()

Good folder hiearchy

Project.Rproj
Data/
Scripts/
Output/

Knitr

Rpubs

R Markdown

Subsetting

X[,1:5] # all rows for columns 1 to 5
X[1:5,] # rows 1 to 5 for all columns
X[(x$var1 <= 3), ] # All rows where value at column var1 is <= 3
 
# Deal with missing values
X[which(X$var2 > 8), ] # return indices are greater than 8, does not return Na's
 
# Sort
sort(X$var1)
X[order(X$var2, X$var3), ] # order by var2, then var3
 
# Ordering with plyr
 
 
# Adding rows and columns
X$var4 <- rnorm(5)
Y <- cbind(X, rnnom(5)) # add a column to the right hand side of X
 
 
# Check for missing values
sum(is.na(X$var1)) # number of missing values in column var1
any(is.na(X))
colSums(is.na(X)) # number of Na's per column
all(colSums(is.na(X))) # number of Na's in entire data set

Cross tabs - similar to x table in Excel

> data("UCBAdmissions")
> DF <- as.data.frame(UCBAdmissions)
> summary(DF)
      Admit       Gender   Dept       Freq      
 Admitted:12   Male  :12   A:4   Min.   :  8.0  
 Rejected:12   Female:12   B:4   1st Qu.: 80.0  
                           C:4   Median :170.0  
                           D:4   Mean   :188.6  
                           E:4   3rd Qu.:302.5  
                           F:4   Max.   :512.0  
> xt <- xtabs(Freq ~ Gender + Admit, data = DF)
> xt
        Admit
Gender   Admitted Rejected
  Male       1198     1493
  Female      557     1278

Flat Tables

Graph View

Backlinks