GEAR: Generic, Efficient, Accurate kNN-based Regression

Keywords: Regression, prediction, k-nearest neighbours, Generic, AccurateAbstract: Regression algorithms are used for prediction (including forecasting of time-series data), inference, hypothesistesting, and modeling of causal relationships. Statistical approaches although popular, are are not generic inthat they require the user to make an intelligent guess about the form of the regression equation. In thispaper we present a new regression algorithm GEAR – Generic, Efficient, Accurate kNN-based Regression. Inaddition to this, GEAR is simple and outlier-resilient. These desirable features make GEAR a very attractivealternative to existing approaches. Our experimental study compares GEAR with fourteen other algorithmson five standard real datasets, and shows that GEAR is more accurate than all its competitors.