Comparison of Density Estimation Methods for Astronomical Datasets

We study the performance of four density estimation techniques. density estimators are applied to six artificial datasets (ad 1-6) and on two astronomical datasets (mgs 1 and 2) derived from the Millennium galaxy sample(mgs) using a Monte Carlo process. We compared the performance of the methods in two ways: first, by measuring the mean squared error and Kullback--Leibler divergence of each of the methods; second, by the visualization of density fields. The results show that the adaptive kernel based methods perform better than the other methods in terms of calculating the density properly.