By: adminPosted on

 Introductions

 

As the concept of big data being broadly introduced to our workplace, measuring the economy of a country has become surprisingly creative. One interesting economic indicator is Lipstick Index, coined by Leonard Lauder, who claims the purchases of lipstick could reflect the health of the economy.

 

Inspired by this concept, we are keen to find the pattern between movies and the U.S. economy. As a matter of fact, the movie industry has been a key driver of the United States economics for a long time, and the film entertainment also becoming a huge part of society. According to the U.S. Bureau of Economic, “Hollywood account for about 504 billion dollars, which is at least 3.2 percent of U.S goods and services.”More importantly, according to the analysis from Bureau of Labor, “Film and TV industry has accounted for supporting 2.1 million jobs and 400,000 local businesses across the country”.

 

 

Therefore, as our primary goal of this project, we intend to find out the relationship between movie production and the U.S economy. More specifically, we are interested in how the types of movie relate to the U.S economy, in other words, which type of movie can realize greater revenue in a specific economic situation. For example, one of the possible findings is that comedy will gain greater revenue during the economy depression.

 

Ultimately, we want our analyses to generate great insights and conduct useful references to movie production companies in the U.S, to help them allocate their assets and investments accordingly to the trend of the domestic economy.

 

 

Data

 

  1. Economic indexes (1959-2018, monthly data)：

inflation rate，personal consumption expenditure, real disposable personal income, real retail and food services sales, Treasury constant maturity rate 10 year

 

  1. Movie stock indexes(1995-2018,daily data):

FOX，IMAX，MSG，DIS，OLED，PGRE，TWX，DISCA

 

  1. Movie indexes

Title, release date, adult, revenue, popularity, genre, overview

 

Movie genre：

       ‘Crime’, ‘Adventure’, ‘Animation’, ‘Comedy’, ‘Drama’, ‘Mystery’,

       ‘Action’, ‘Western’, ‘Science Fiction’, ‘Thriller’, ‘Romance’,

       ‘Documentary’, ‘Fantasy’, ‘Horror’, ‘Music’, ‘History’, ‘War’,

       ‘Family’, ‘TV Movie’

 

Data processing

  1. Removing NA. choosing the period 2003-2013
  2. Use ‘inflation’ to modify attributes related to money
  3. Remove outliers.

E.g. removing movies with ‘revenue’<1000

  1. Transfer all the dataset into monthly data.

E.g. for the daily dataset, take the average value of this attribute in a specific month to represent its value on month level

  1. Using average stock prices of the listed movie companies to represent the whole movie stock market