# R-help to Exercise 3

�

# Read the data into R and look at the data:

speed=scan("http://www.uio.no/studier/emner/matnat/math/STK4900/v26/exercises/exer2.dat")

speed

�

�

# QUESTION a)

�

# Plot the data in various ways:

hist(speed)��� ��������������� # histogram

plot(ecdf(speed)) �����������#empirical distribution function

boxplot(speed)� ���������� # box plot

�

# What do the different plots tell you?

# Are there indications for "outliers" in the data?

�

�

# QUESTION b)

�

# Compute the (empirical) mean and the median, which are the two most common measures of location

# (alternatively you may use the command summary(speed) to compute the mean and the median)

mean(speed)���������������� #mean

median(speed)������������� #median

�

# What do the two measures of location tell you?

# QUESTION c)

# Compute the (empirical) standard deviation and the interquartile range, which are two common measures of spread:

sd(speed)�������������������� # standard deviation

IQR(speed)����������������� # interquartile range

�

# What do the two measures of spread tell you?

�

�

# QUESTION d)

�

# Compute t-based confidence interval using all data:

t.test(speed)

�

# Compute t-based confidence interval without the two "outliers":

t.test(speed[speed>0])

�

�

# What do the two intervals tell you? Which one do you find most reasonable?