# ---------------------------------------------------------------------
# Program: Bivariate1.S  
#  Author: Steven M. Boker
#    Date: Tue Oct 3 08:10:36 EDT 2006
#
#  Bivariate data
#
# ---------------------------------------------------------------------

# ----------------------------------------------
#  Load required libraries

library(lattice)

# ------------------------------------------------
# Read the iris data. 

iris <- data.frame(read.table("iris.dat", header=T))
summary(iris)

source("edaStudent.sdd")
summary(edaStudent)

# ---------------------------------------------------------------------
#  Scatterplots


pdf("Biv1ZtestPlot1.pdf", height=5, width=5)
plot(edaStudent$ztest, edaStudent$hscgpa, 
     xlab="Entrance Exam (Z-Score)",
     ylab="High School Core GPA",
     type='p',
     pch=2)
dev.off()

pdf("Biv1HSCGPAXYPlot1.pdf", height=5, width=5)
print(xyplot(hscgpa ~ ztest,
        data = edaStudent,
        aspect=1,
        xlab = list(label="Entrance Exam (Z-Score)",cex=1.25),
        ylab=list(label="High School Core GPA",cex=1.25),
        scales=list(cex=1.25)
        )
    )
dev.off()

pdf("Biv1ZtestXYPlot0.pdf", height=5, width=5)
print(xyplot(ztest ~ sex,
        data = edaStudent,
        aspect=1,
        xlab = list(label="Sex",cex=1.25),
        ylab=list(label="Entrance Exam (Z-Score)",cex=1.25),
        scales=list(cex=1.25)
        )
    )
dev.off()


pdf("Biv1ZtestXYPlot1.pdf", height=5, width=5)
print(xyplot(ztest ~ jitter(as.numeric(sex), .5),
        data = edaStudent,
        aspect=1,
        xlab = list(label="Sex",cex=1.25),
        ylab=list(label="Entrance Exam (Z-Score)",cex=1.25),
        scales=list(cex=1.25)
        )
    )
dev.off()


# ---------------------------------------------------------------------
#  Scatterplots conditioned on a factor


pdf("Biv1HSCGPAXYPlot2.pdf", height=5, width=6)
print(xyplot(hscgpa ~ ztest | sex,
        data = edaStudent,
        aspect=1,
        xlab = list(label="Entrance Exam (Z-Score)",cex=1.25),
        ylab=list(label="High School Core GPA",cex=1.25),
        scales=list(cex=1.25)
        )
    )
dev.off()


# ---------------------------------------------------------------------
#  Scatterplots conditioned on a factor


pdf("Biv1IrisPlot1.pdf", height=5, width=5)
print(xyplot(petal.length ~ petal.width | variety,
        data = iris,
        aspect=1,
        xlab = list(label="Petal Length",cex=1.25),
        ylab=list(label="Petal Width",cex=1.25),
        scales=list(cex=1.25)
        )
    )
dev.off()


pdf("Biv1IrisPlot2.pdf", height=5, width=5)
print(xyplot(sepal.length ~ sepal.width | variety,
        data = iris,
        aspect=1,
        xlab = list(label="Sepal Length",cex=1.25),
        ylab=list(label="Sepal Width",cex=1.25),
        scales=list(cex=1.25)
        )
    )
dev.off()


# ---------------------------------------------------------------------
#  Scatterplots conditioned on a continuous variable

pdf("Biv1HSCGPAXYPlot3.pdf", height=5, width=6)
print(xyplot(hscgpa ~ ztest | cut(pposths,6),
        data = edaStudent,
        aspect=1,
        main=list("HSCGPA vs. ZTEST \n Cut by Probability of Post HS in Neighborhood",cex=1.25),
        xlab = list(label="Entrance Exam (Z-Score)",cex=1.25),
        ylab=list(label="High School Core GPA",cex=1.25),
        scales=list(cex=1.25)
        )
    )
dev.off()

# ---------------------------------------------------------------------
#  Estimating a linear model

tLM <- lm(hscgpa ~ ztest, data=edaStudent)
summary(tLM)

# ---------------------------------------------------------------------
#  Plotting a regression line

pdf("Biv1HSCGPAXYPlot4.pdf", height=5, width=5)
print(xyplot(hscgpa ~ ztest, 
        data = edaStudent,
        panel = function(x, y) {
            panel.xyplot(x, y)
            panel.abline(lm(hscgpa ~ ztest, data = edaStudent))
        },
        aspect=1,
        xlab = list(label="Entrance Exam (Z-Score)",cex=1.25),
        ylab=list(label="High School Core GPA",cex=1.25),
        scales=list(cex=1.25)
        )
    )
dev.off()