Rules & FAQs
Rules
The rules for Analytics X are as follows:
1.) Only one account is allowed per person.
2.) A user may submit only one entry every 1 day(s).
3.) Submissions are scored using RMSE, a method for judging the accuracy of predictions.
4.) Submissions are not eligible for awards until they have been tested against at least 25 data points. The Score (RMSE) is calculated by comparing the user's submission to homicides that occur after the submission upload date.
5.) A Monthly Progress Prize ($20) is awarded to the best performing submission at the end of each month. A Grand Prize ($100) is awarded at the end of the calendar year to the best performing submission.
6.) In the event of a tie, the award is given to the earlier submitted entry. Further ties cause an award to be split evenly among the best submissions.
Rules are subject to change.
FAQ
What is the purpose of this website?
Analytics X is an ongoing contest to apply analytics, modeling, and statistics to solve the social problems that affect our cities. It combines the fields of statistics, mathematics, and social science to understand the root causes of dysfunction in our neighborhoods. Understanding these relationships and discovering the most highly correlated variables allows us to deploy our limited resources more effectively and target the variables that will have the greatest positive impact on improvement.
How does it work?
Contestants predict the fraction of all 2010 Philadelphia homicides that
will occur in each of the 47 major zip codes. These predictions are
uploaded (using the Submit a Prediction button) and are then automatically
compared to actual homicides that occur throughout the year. The best
performing models are given awards at the end of each month. The
submission must be a tab delimited text file (.txt) with only two columns
of data. The first column must contain all 47 major 5 digit zip codes in
Philadelphia. The second column must be a number that indicates the
fraction of all Philadelphia homicides that will occur in that zip code in
2010. Ex.) If your model predicts that half of all Philadelphia homicides
will occur in 19111, then 19111 should have a 0.5 next to it. You can
download a sample file here.
Which homicides are being modeled in this competition?
Unlawful homicides where there was intent to kill are being modeled in Analytics X.
I like thinking about this stuff and I'm pretty good at math; Can I make a career out of this kind of thing?
Yes, you've just described the field of Operations Research and Decision Science. Operations Research is one of the fastest growing career fields because it combines multiple fields of study and can offer solutions to very complex problems.
Why are the prize amounts so low?
Analytics X is privately funded and has no sponsors. People who like working on these kinds of problems generally do so because they find them interesting.
