Apartment

mydata <- read.table("./Apartment.csv", 
                     header=TRUE, 
                     sep=";", 
                     dec=",")

head(mydata)

##   ID Price
## 1  1  3144
## 2  2  3512
## 3  3  3555
## 4  4  3322
## 5  5  2980
## 6  6  3732

Description:

• ID: ID of apartment
• Price: Price per m2.

mean(mydata$Price)

## [1] 3109.677

sd(mydata$Price)

## [1] 476.681

library(ggplot2)
ggplot(NULL, aes(c(-4, 4))) +
  geom_line(stat = "function", fun = dt, args = list (df = 30)) +
  ylab("Density") + 
  xlab("Sample estimates") +
  labs(title="Distribution of sample estimates")

qt(p = 0.025, df = 30, lower.tail = FALSE)

## [1] 2.042272

qt(p = 0.025, df = 30, lower.tail = TRUE)

## [1] -2.042272

t.test(mydata$Price,
       mu = 2770,
       alternative = "two.sided")

## 
##  One Sample t-test
## 
## data:  mydata$Price
## t = 3.9675, df = 30, p-value = 0.0004175
## alternative hypothesis: true mean is not equal to 2770
## 95 percent confidence interval:
##  2934.829 3284.525
## sample estimates:
## mean of x 
##  3109.677

Based on the sample data we reject the H0 at p < 0.001. we find that the average price increased.

Calculation of effect size

Cohen’d statistics

#install.packages("effectsize")
library(effectsize)

effectsize::cohens_d(mydata$Price, mu=2770)

## Cohen's d |       95% CI
## ------------------------
## 0.71      | [0.31, 1.10]
## 
## - Deviation from a difference of 2770.

effectsize::interpret_cohens_d(0.71, rules = "sawilowsky2009")

## [1] "medium"
## (Rules: sawilowsky2009)