Probability Density Estimation of stochastic electric load is of most importance nowadays in power system operations and urban planning. This is due to the continuous demand to integrate intermittent renewable energy resources that introduce uncertainties in the operating state of power systems which in turn requires accurate and reliable methods to estimate load. This paper is the first to employ a nonparametric techniques called Root Transform Local Linear Regression for estimating electric load. This robust model proposed estimates electric load data  more accurately than parametric models used in current literature. The performance of the root transform local linear regression model is compared with two kernel density estimation models and two parametric models (Gaussian and Gamma distributions) and is assessed using the Kolmogorov-Smirnov goodness-of-fit test, Coefficient of determination and four error metrics. Results confirm the accuracy of the nonparametric models over the parametric models with the root transform model performing best across all error metrics and K-S test, followed by the kernel density estimation model. An interactive web application is developed to perform the same analysis presented in this paper on any type of univariate data.