Lecture13.ipynb

Kernel: R (R-Project)

Lecture 13: Boostrap and Confidence Intervals

• Mean, Median, Mode, Percentiles

• The Bootstrap Method; Confidence Intervals

mylist4 <- c( 75, 57, 75, 81, 89, 100)
sort(mylist4)

1. 57
2. 75
3. 75
4. 81
5. 89
6. 100

quantile( mylist4, 0.6, type = 1 )

60%: 81

library('dplyr')
library('ggplot2')

Attaching package: ‘dplyr’

The following objects are masked from ‘package:stats’:

    filter, lag

The following objects are masked from ‘package:base’:

    intersect, setdiff, setequal, union

payroll <- read.csv( 'citywidepayroll_small.csv' )

original_sample <- sample( payroll$Regular.Gross.Paid, 100, replace = FALSE )

bootstrap_samples <- data.frame( ave_RGP = double(2000) )

count <- 1

while( count <= 2000){
    sample_group <- sample( original_sample, 100, replace = TRUE )
    bootstrap_samples$ave_RGP[count]  <- mean( sample_group )
    
    count <- count + 1
}

ggplot(  bootstrap_samples, aes( x = ave_RGP )) + geom_histogram( bins = 20 )

mean( bootstrap_samples$ave_RGP)

49001.02280295

true_ave <- mean( payroll$Regular.Gross.Paid)
true_ave

39952.342948