team                            DATE                             I LEARNED                    TOOLS    

Lisa Cheng                                                                    Fall 2015                                                                        SQL - Join                                                                    SAS Stuido

Michel Zou                                                                                                                                                           ANOVA

Sarvari Ventrapragada                                                                                                                                        Logistic Regression

                                                                                                                                                                             ROC Curve

#Does hollywood pick the right movie to produce? 

Hollywood film executives often wrestle with the decision of whether or not to back a film with blockbuster potential. It may seem promising if hot directors love the script and major stars are interested in it, too. But in Hollywood, there are no guarantees. That project you're thinking of green lighting could wind up being a hit - or it could be a career-destroying flop. With films often costing hundreds millions and up, film studios need a better way than gut feelings to ensure they invest in the right movie.

How can film executives predict the next blockbuster correctly? Is there any more accurate methods to predict a successful movie, rather than filmmaker's intuitions?   

#We aim to help film executives use analytics to predict the next blockbusters!

1. How to define a blockbuster? what factors will impact a successful movie?

2. how to use analytics to predict a successful movie?

#WE merged data sets from imdb.com and boxofficemojo.com, and conducted Exploratory Analysis to get feel of our movie data.

1. statistic summary of merged data sets

2. EXPLORATORY ANALYSIS

There are some very interesting patterns behind the our merged data set which includes predictors such as production company, directors, budget, sequel. For example,  industry market leaders such as Walt Disney are not the top profitable film companies. While,Dune Entertainment, which has produced some very famous films like Night at the Museum and X-men, becomes the most profitable film company after merging with a famous American director and becoming Warner Bros' key co-financing partner.  Not surprisingly, an expert in super hero film, Marvel Entertainment, also is one of the quickest money-makers. 

A good box office collection doesn't mean that this film maker invested in a right movie in terms of shareholders' perspectives. Also, among well-known directors, some directors turns film into money very quickly. Like one of my favorite directors, Christopher Nolan, he usually doesn't spend too much on his film but most of his films - I cannot wait not to mention Batman series, occupied all seats in theaters. While some created huge box offices, we are not sure how much they spent on producing those movies. As one of our goals-predicting a profitable movie, we not only focus on box office collection but also consider ROI (Return on Investment) of those movies. 

3. Correlation 

The bigger movie budget, the huger box office collection? Yes! At least there is a high correlation between total film box office and its budget within our data sets. 

But there is no clue that a bigger budget lead to a high-rated movie. 

# Our  secret equations of blockbusters.

Total Box Office=2.37*budget + 153*sequel +194*runningtTime(long)+ 107*creativeType (Kids Fiction) + 100*creativeType(Science Fiction)+99*Boardrating(G)

IMDB Rating= 6.26+0.2*Director(High Productive) +1*RunningTime(long)+0.56*Creative Type(Dramatization)+ 0.49*Creative Type(Historical Fiction)- 0.35*BoradRating(PG-13)