--- title: "Checking for independence - a simple KnitR example" author: "Martin Papke" date: "23 August 2018" output: pdf_document --- ```{r setup, include=FALSE} knitr::opts_chunk$set(echo = TRUE) library(knitr) library(readr) ``` # A simple KnitR example ## Data import In this document we aim to show how KnitR can be used to gerenate a report or an article containing statistical data and how the R code can be integrated within the document. As example data, we use 10000 dice rolls contained in the file *dice.csv*. As usual in R we can load the data with ```{r loaddata} # data <- read.csv('dice.csv', stringsAsFactors=FALSE) # dice <- as.numeric(data$X3) ``` To give a standalone example here, we use R's feature to generate random numbers ``` dice <- sample(1:6, 10000, replace=TRUE) ``` ### Preperation of the data As we want to use the dice throws in pairs, we just generate a table comparing the even and the odd dice throws, this can be done as follows: ```{r dataprep} even <- dice[seq.int(0,10000,2)] odd <- dice[seq.int(1,10000,2)] tbl <- table(even, odd) ``` We obtain the results ```{r table1, echo=FALSE} kable(tbl, caption='even and odd results compared') ``` ## Statistics No we check for independence, by invoking ```{r chisquared} chi <- test.chisq(tbl) p <- chi$p.value ``` The $p$-value is `r p`. Hence, we can say that we have ```{r pvalue} if (p < 0.01) { "high significance for independene" } else if (p < 0.05) { "significance for independence" } else { "no significance for independence" } ```