Handout 7

Please let us use this handout to discuss Associations /Correlations. I would like you to make comments as you run the code we have listed in this handout. In your submission file you should include commments that show a clear understanding of the correlation concept.

par() function is used to set or query graphical parameters.

# Create variable mfrow is a vector of the form c(nr, nC) to draw in this case the arrays on the device by rows.
mfrow <- par()$mfrow

# Create a variable mar using the par() function and using mar which gives the number of lines of margin to be specified on the four sides of the plot.
mar <- par()$mar

# Create a varibale oma using the par() function and usin oma to give the size fo the outer margins in lines of text.
oma <- par()$oma

# Using the par function to combine the all the graphical parameters from the using the defined variable above.
par(mfrow=c(2, 3), mar=c(0, 0, 2, 0), oma=c(1, 1, 1, 1))

# set.seed creates a replicate of the random generation two numbers.
set.seed(2)

# Creates a uniform distribution of 500 numbers between 0 and pi
x <- runif(500, min=0, max=pi)

# Create a variable y to hold the rnorm density, distribution function, quantile function and random generation for the normal distribution with mean equal to the x and the standard deviation equal to 0.2.
y <- rnorm(500, mean=x, sd=0.2)

# Create a plot diagram using a vector plotting characters in this case 19 is a solid circle, cex is the numeric vector giving the amount by which plotting characters and symbols should be scaled relative to the default, col is the color, axes is whether to draw a axes on the plot in this case it is FALSE, xlab and ylab are labels for the x axis and y- axis.
plot(x,y, pch=19, cex=.8, col="#666699CC", axes=FALSE, xlab="", ylab="", 
     main=paste("Designation: ", round(cor(x,y), 2)))

Make comments about the output as well as the syntax related to the chunk ran above. # I made the comments in the code. # The diagram showing a correlation the points are closely packed and as x increases to does y. A simple linear regression.

y <- rnorm(500, mean=-x, sd=0.3)
plot(x,y, pch=19, cex=.8, col="#666699CC", axes=FALSE, xlab="", ylab="", 
     main=paste("Correlation: ", round(cor(x,y), 2)))

y <- rnorm(500, mean=sin(x), sd=0.2)
plot(x,y, pch=19, cex=.8, col="#666699CC", axes=FALSE, xlab="", ylab="", 
     main=paste("Correlation: ", round(cor(x,y), 2)))

y <- rnorm(500, mean=x, sd=2)
plot(x,y, pch=19, cex=.8, col="#666699CC", axes=FALSE, xlab="", ylab="", 
     main=paste("Correlation: ", round(cor(x,y), 2)))

y <- rnorm(500, mean=-x, sd=1)
plot(x,y, pch=19, cex=.8, col="#666699CC", axes=FALSE, xlab="", ylab="", 
     main=paste("Correlation: ", round(cor(x,y), 2)))

y <- runif(500, min=0, max=pi)
plot(x,y, pch=19, cex=.8, col="#666699CC", axes=FALSE, xlab="", ylab="", 
     main=paste("Correlation: ", round(cor(x,y), 2)))