Predictive modeling is a statistical data mining approach that builds a prediction function from the observed data. The function is then used to estimate a value of a dependent variable for new data. A commonly used predictive modeling method is regression that has been applied to a wide range of application domains. Inthispaper,webuildmultivariateregressionmodelsofhomepricesusingadataset composed of 81 homes. We then applied the maximum information coefficient (MIC) statistics to the observed home values (Y) and the predicted values (X) as anevaluationoftheregressionmodels.Theresultsshowedveryhighstrengthofthe relationship between the two variables X andY.
Key words: Predictive modeling, multivariate linear regression, hedonic price model, maximum information coefficient.