The Delta Method is a useful approach for estimating expectation and variances of the function of a random variable. This method makes use of the Taylor series approximation of the mean and variance of a random variable. It is particularly useful for functions of asymptotically normal statistical estimators. One use case it is used in machine learning is to estimate the variance of the estimator used in the estimation of parametric binormal ROC curves.

Delta Method

--

--

Nuzhi Meyen

Co-founder of Helios P2P. Sri Lankan. Interested in Finance, Advanced Analytics, BI, Data Visualization, Computer Science, Statistics, and Design Thinking.