I am now working on project related to statistic.
So I am not expertise in this field actually.
I would like to ask you guys how to predict the revenue of new movie for the first week revenue in box office from IMdb data.
Which statistical model is suitable for example linear regression or whatever?
What data should I collect to start up?
Thank you
Perhaps the best method would be to find a similar move from the past (not to distant), and observe that movies first week revenue if data is available. Then some questions related to the differences of these two movies should be asked and answered. These answers should then be used to try to state whether the film at hand has potential to sell more or less tickets the first week compared to the (historic) film you have picked to compare with.
I don't think there is a specific model ... or someone had a similar idea and published an article about it with little or no recognition.
I would first (as beforementioned) collect data about movie releases of different countries (if available). I guess there will be at least some data for developed countries. Either in magazines or the publishers of the movies have these information. Then you can maybe make an OLS regression with other valuables like population variables (because I would think that people between the age of 9 or 16 and 29 are more probable to watch a movie as soon as possible), then disposable income in that country, average price of a movie ticket, and so on. Important as another regressor would be, if the movie is a "sequel" or a "prequel", a "remake", (dummy variables), what is the genre of the movie? Famous actors (difficult to measure)?... and then you can build it up by yourself.
If that's too difficult, ask companies like Netflix, Amazon, Disney, HBO,... because they are using Big Data now.
Maybe this paper can be useful for you...
https://link.springer.com/article/10.1007/s11747-014-0388-3
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude