Title:A new class of binning free, multivariate goodness-of-fit tests: the energy tests

Abstract: The energy of two charge distributions of opposite total charge is minimum when their distributions are equal. This property has been used to define a test quantity measuring the compatibility of an experimental data sample with a null hypothesis $H_{0}$. The pdf corresponding to $H_{0}$ is simulated by discrete elements, the simulation charges. The test can be applied to multivariate samples. The test quantity called ''energy'' depends on a correlation function of the distances of the charges. It can be adjusted to the expectation of possible deviations of the sample from H0. We have investigated the power of the test for a uniform and a Gaussian distribution in one and two dimensions, respectively and compared it to that of conventional tests. The energy test with a Gaussian correlation function is closely related to the Pearson chi squared test but is more powerful in most applications and avoids arbitrary bin boundaries. A logarithmic correlation function is recommended for the detection of long range distortions which frequently occur in physics applications.