Esm
(2IAB0) Graded Assignment 2: Movies 1332
Introduction
The success of movies depends on different factors. Not only the genre, duration and release month (in
are important, but also the Holidays and whether or not it is an Oscar-winning movie can have an effect
will be on Valentine’s Day; does Valentine’s Day have a positive effect on the daily revenue of movies? A
investigated are Oscars; Are Oscar-winning movies more likely to be profitable than non-Oscar-winning
Method for 1st question Method for 2nd ques
To answer this question, it is important to first combine the movies For the second question, the
dataset and daily dataset into one data frame. Out of this combined combined. Then there is a co
data frame, only the movies showed on and around Valentine’s Day the data frame has been spl
are selected (from 1 up to 15 February). Then, this Valentines data winning movies and one with
frame is grouped by genre and the average gross of each genre is the question the hypothesis h
calculated. This is also done for all genres showed over the whole year. hypothesis is that the mean p
These two average grosses for each genre are plotted in one figure. In movies are the same. The al
this figure can be seen if the average daily gross is indeed higher profit of winning movies is lar
during the Valentine’s period than during the whole year. What also can movies. To check for normal
be seen in the figure is which genre in particular has a higher gross or Anderson-Darling normality t
even which genre has a lower gross during the Valentine’s period. further check for normality, th
Finally, it is also important to look at the day of the week. Therefore, the the log of the data. When the
average gross for each day of the week is plotted. reject the null hypothesis and
Results for 1st question Results for 2nd ques
Figure 1 shows the average daily gross for each genre during the Figure 4 shows the
whole year and during the Valentine’s period (02/01-02/15). Q-Q plots, the points
do not really follow the
line, which indicates
that there is no normal
distribution.
The Anderson-Darling
normality test results
in a p-value of 1.7e-29
for winning movies
and 0.0 for non-
winning movies. These
values are both lower
than 0.05. This again
Figure 1: Average daily gross indicates that there is
for each genre
no normal distribution.
In this figure it is not really obvious if the daily gross is higher during the
Valentine’s period, therefore the actual means are calculated. The
Figure
mean of daily gross during the Valentine’s period is 1.040.421, while
during the whole year this value is 929.318. This means that there is a
To check, the Q-Q
rather small difference.
plots also have been
Because of this small difference, the above has been done again. This made for the log of the
time not the Valentine’s period is considered, but only Valentine’s Day data. As can be seen
itself. Figure 2 shows the average daily gross for each genre during the in figure 5, there is still
whole year and on Valentine’s Day. This figure shows a clear no sign of normality.
difference. The average daily gross for each genre is much larger on This means that we
Valentine’s Day than during the whole year. can reject the null
hypothesis, which
means that the alter-
native hypothesis is
correct.
(2IAB0) Graded Assignment 2: Movies 1332
Introduction
The success of movies depends on different factors. Not only the genre, duration and release month (in
are important, but also the Holidays and whether or not it is an Oscar-winning movie can have an effect
will be on Valentine’s Day; does Valentine’s Day have a positive effect on the daily revenue of movies? A
investigated are Oscars; Are Oscar-winning movies more likely to be profitable than non-Oscar-winning
Method for 1st question Method for 2nd ques
To answer this question, it is important to first combine the movies For the second question, the
dataset and daily dataset into one data frame. Out of this combined combined. Then there is a co
data frame, only the movies showed on and around Valentine’s Day the data frame has been spl
are selected (from 1 up to 15 February). Then, this Valentines data winning movies and one with
frame is grouped by genre and the average gross of each genre is the question the hypothesis h
calculated. This is also done for all genres showed over the whole year. hypothesis is that the mean p
These two average grosses for each genre are plotted in one figure. In movies are the same. The al
this figure can be seen if the average daily gross is indeed higher profit of winning movies is lar
during the Valentine’s period than during the whole year. What also can movies. To check for normal
be seen in the figure is which genre in particular has a higher gross or Anderson-Darling normality t
even which genre has a lower gross during the Valentine’s period. further check for normality, th
Finally, it is also important to look at the day of the week. Therefore, the the log of the data. When the
average gross for each day of the week is plotted. reject the null hypothesis and
Results for 1st question Results for 2nd ques
Figure 1 shows the average daily gross for each genre during the Figure 4 shows the
whole year and during the Valentine’s period (02/01-02/15). Q-Q plots, the points
do not really follow the
line, which indicates
that there is no normal
distribution.
The Anderson-Darling
normality test results
in a p-value of 1.7e-29
for winning movies
and 0.0 for non-
winning movies. These
values are both lower
than 0.05. This again
Figure 1: Average daily gross indicates that there is
for each genre
no normal distribution.
In this figure it is not really obvious if the daily gross is higher during the
Valentine’s period, therefore the actual means are calculated. The
Figure
mean of daily gross during the Valentine’s period is 1.040.421, while
during the whole year this value is 929.318. This means that there is a
To check, the Q-Q
rather small difference.
plots also have been
Because of this small difference, the above has been done again. This made for the log of the
time not the Valentine’s period is considered, but only Valentine’s Day data. As can be seen
itself. Figure 2 shows the average daily gross for each genre during the in figure 5, there is still
whole year and on Valentine’s Day. This figure shows a clear no sign of normality.
difference. The average daily gross for each genre is much larger on This means that we
Valentine’s Day than during the whole year. can reject the null
hypothesis, which
means that the alter-
native hypothesis is
correct.
