Demonstrating bi-variate correlations using simulation.

server.R

# Script to demonstrate distributions
library(eeptools)
library(shiny)
library(ggplot2)


rnormcor <- function(x,rho) rnorm(1,rho*x,sqrt(1-rho^2))

shinyServer(function(input,output){
  output$distPlot<-reactivePlot(function(){
    a<-rnorm(input$obs)
    b<-sapply(a,rnormcor,rho=input$rho)
    p<-qplot(a,b,alpha=0.85)+geom_smooth(method="lm",se=FALSE,size=1.1)+theme_dpi()
    p<-p+labs(x="",y="",title="Demonstrating Correlations")
    p<-p+geom_text(aes(x=-2.5,y=3,label=paste("Corr. =",input$rho,sep=" ")),size=8)
    print(p)
    
  })
})

ui.R

# Script to demonstrate distributions
library(eeptools)
library(shiny)
library(ggplot2)
shinyUI(pageWithSidebar(
  
  # Title
  headerPanel("Simulating Data with Correlation"),
  
  sidebarPanel(
    sliderInput("obs","Number of observations:",
                min=200,max=5000,value=500,step=250),
    sliderInput("rho","Correlation Coefficient",
                min=-1,max=1,value=0,step=0.1)    
  ),
  
  # GGPLOT
  
  mainPanel(
    plotOutput("distPlot")
  )
  
  
  
))