Description:

Before asking someone on a date or skydiving, it's important to know your likelihood of success. The same goes for quoting home insurance prices to a potential customer. Homesite, a leading provider of homeowners insurance, does not currently have a dynamic conversion rate model that can give them confidence a quoted price will lead to a purchase. 

Using an anonymized database of information on customer and sales activity, including property and coverage information, Homesite is challenging you to predict which customers will purchase a given quote. Accurately predicting conversion would help Homesite better understand the impact of proposed pricing changes and maintain an ideal portfolio of customer segments. 

Evaluation:

Submissions are evaluated on area under the ROC curve between the predicted probability and the observed target.

Submission File

For each QuoteNumber in the test set, you must predict a probability for QuoteConversion_Flag. The file should contain a header and have the following format:

QuoteNumber,QuoteConversion_Flag
3,0
5,0.3
7,0
etc.

Data Description:

This dataset represents the activity of a large number of customers who are interested in buying policies from Homesite. Each QuoteNumber corresponds to a potential customer and the QuoteConversion_Flag indicates whether the customer purchased a policy.

The provided features are anonymized and provide a rich representation of the prospective customer and policy. They include specific coverage information, sales information, personal information, property information, and geographic information. Your task is to predict QuoteConversion_Flag for each QuoteNumber in the test set.

File descriptions

• train.csv - the training set, contains QuoteConversion_Flag

• test.csv - the test set, does not contain QuoteConversion_Flag

• sample_submission.csv - a sample submission file in the correct format