Nonparametric statistics encompasses the statistical methods that do not make any assumptions about the underlying distribution of data. It can be contrasted with parametric statistics, which makes explicit assumptions about the distribution of observed data and which uses the data to estimate parameters of that distribution. For example, it is common to assume that data is drawn from a normal distribution with unknown parameters μ (mean) and σ2 (variance); a parametric approach would estimate μ and σ by the sample mean and the sample standard deviation, respectively. Nonparametric statistics, on the other hand, would not assume the data is normally distributed a priori, and instead would estimate the shape of the distribution itself.
There are two main types of nonparametric methods in statistics: methods that attempt to discover the unknown underlying distribution of the data, and methods that make statistical inferences without...
References and Readings
- Tamhane, A. C., & Dunlop, D. D. (2000). Statistics and data analysis: From elementary to intermediate (pp. 562–604). Upper Saddle River, NJ: Prentice-Hall.Google Scholar
- Wasserman, L. (2004). All of statistics (pp. 303–326). New York: Springer.Google Scholar
- Wasserman, L. (2006). All of nonparametric statistics. New York: Springer.Google Scholar