dsparks

is.R()  # Should return TRUE if run in R.

dsparks

# Let's say you want to display a table of model coefficients
# in order of their significance,
# and you want to plot the distribution of model residuals,
# but you don't know how to access these values.
# Use str().

# Generate some random data
NN <- 1000
theData <- data.frame(Alpha = rnorm(NN),
                      Beta = rnorm(NN))

dsparks

# Let's say you have an R object that you'd like to share with someone else
# but, for whatever reason, it is necessary to share it in text form.

# Generate a random data.frame
set.seed(1337)
NN <- 10
theData <- data.frame(Alpha = rnorm(NN),
                      Beta = rnorm(NN))
theData$Gamma <- theData$Alpha * 2 + theData$Beta / 2 + rnorm(NN)
print(theData)

dsparks

# A method for modifying only select off-diagonal items in a matrix
# From "Thierry" and "Ben Bolker"
# At http://stackoverflow.com/a/11759744/479554

# A sample matrix
size <- 6
mat <- matrix(seq_len(size ^ 2), ncol = size)
print(mat)

# A companion matrix that indicates how "off" a diagonal is:

dsparks

# Randomly allocating observations into groups, for, e.g. cross-validation
kk <- 10  # Number of partitions, as in "kk-fold cross-validation."

# Here is a data.frame full of good data:
nn <- 1003
myData <- data.frame(matrix(rnorm(nn * 3), ncol = 3))
colnames(myData) <- LETTERS[1:3]

# This does not work:
whichK <- sample(LETTERS[1:kk], nrow(myData), replace = T)

dsparks

# Alternative to a doubly-nested loop
# Imagine I want to perform an operation on a data frame
# once for each combination of two variables, such as Country and Year
# I can do this with a nested loop, or I can do this with (among other
# things) lapply()

# Generate random data:
allCountries <- LETTERS[1:10]
allYears <- 1990:2012

dsparks

# A simple approach to visually-weighted regression plots

doInstall <- TRUE  # Change to FALSE if you don't want packages installed.
toInstall <- c("ggplot2", "reshape2", "MASS")
if(doInstall){install.packages(toInstall, repos = "http://cran.us.r-project.org")}
lapply(toInstall, library, character.only = TRUE)

# Generate some data:
nn <- 1000
myData <- data.frame(X = rnorm(nn),

dsparks

doInstall <- TRUE  # Change to FALSE if you don't want packages installed.
toInstall <- c("ggplot2", "reshape2", "RColorBrewer")
if(doInstall){install.packages(toInstall, repos = "http://cran.us.r-project.org")}
lapply(toInstall, library, character.only = TRUE)

# Generate a random matrix
# This can be any type of numeric matrix,
# though we often see heatmaps of square correlation matrices.
nRow <- 9
nCol <- 16

dsparks

# https://github.com/wch/ggplot2/wiki/New-theme-system
new_theme_empty <- theme_bw()
new_theme_empty$line <- element_blank()
new_theme_empty$rect <- element_blank()
new_theme_empty$strip.text <- element_blank()
new_theme_empty$axis.text <- element_blank()
new_theme_empty$plot.title <- element_blank()
new_theme_empty$axis.title <- element_blank()
new_theme_empty$plot.margin <- structure(c(0, 0, -1, -1), unit = "lines", valid.unit = 3L, class = "unit")

dsparks

# Transforming a color scale

doInstall <- TRUE  # Change to FALSE if you don't want packages installed.
toInstall <- c("ggplot2", "RColorBrewer", "scales")
if(doInstall){install.packages(toInstall, repos = "http://cran.us.r-project.org")}
lapply(toInstall, library, character.only = TRUE)

# Generate some data
nn <- 10000
myData <- data.frame(X = rnorm(nn),