Adjusted Overfitting Regression

Abstract: In this paper, I will introduce a new form of regression, that can adjust overfitting and underfitting through, "distance-based regression". Overfitting often results in finding false patterns causing inaccurate results, so by having a new approach that minimizes overfitting, more accurate predictions can be derived. Then I will proceed with a test of my regression form and show additional ways to optimize the regression. Finally, I will apply my new technique to a specific data set to demonstrate its practical value. CONTENTS Introduction 1. Distance and X-axis Based Regression 1.1 X-Axis Based Regression 1.2 Distance Based Regression 2. Weighted Regression 2.1 Division "Weighted Cost Functions" 2.2 Other "Weighted Cost Functions" 2.3 Randomness and change adjusted "Weighted Cost Functions" 3. Applications and Tests 3.1 Testing on Different Data sets References Index Wilson 2 Introduction In this paper I will introduce a new form of regression, "Overfitting Based Regression" which allows you to tune the level of overfitting or underfitting, with the goal of generalizing standard regression methods. This new regression technique produces a nonlinear function of the x or right hand side variables using weights on neighboring data points, instead of the traditional approach of applying the best fit line.