In jrminter/statshelpR: Convenience functions for statistical analysis

knitr::opts_chunk$set(echo = TRUE, warning=FALSE)
suppressWarnings(suppressMessages(suppressPackageStartupMessages(library(ggplot2))))

Introduction

The stats4stem.org website provides a table with the desired confidence level for a $t^*$ calculation and the the confidence level entered in R's qt() function. This is presented without explanation.

This is an exploratory data analysis

Sumarize the data

We can load the data in R as shown below.

library(statshelpR)
library(pander)

data(tstar_cl_r)
pander(tstar_cl_r)

Fit the data to a linear model

Start with a plot

We can plot the data and fit it to a linear model

plt <-  ggplot(tstar_cl_r, aes(x=CL, y=R)) +
        geom_point(color='black', size=2) +
        xlab("CL") + ylab("R") +
        ggtitle("R as a function of CL") +
        theme(axis.text=element_text(size=12),
              axis.title=element_text(size=14),
              plot.title = element_text(hjust = 0.5))
print(plt)

Compute the linear model

Let's do the linear model

fit <- lm(R ~ CL, data = tstar_cl_r)
pander(summary(fit))

Add the line to the plot

And let's update our plot with the predicted line.

pred <- data.frame(R = predict(fit, tstar_cl_r))
pred$CL <- tstar_cl_r$CL
pander(pred)

plt <- plt +  geom_line(color='red',data = pred, aes(x=CL, y=R))
print(plt)

So our equation is $R = 0.5*CL + 0.5$

Make a data frame of calulated values

CL <- tstar_cl_r$CL

CL_to_R <- function(x){
  R <- 0.5 * x + 0.5
  return(R)
}

rVal <- sapply(CL, CL_to_R)

dfCalc <- data.frame(CL=CL, R=rVal)
pander(dfCalc)