VARIABLE GROUPING-BASED SKEWED DISTRIBUTION OPTIMAL PARAMETER ESTIMATION METHOD

摘要

Disclosed in the present invention is a variable grouping-based skewed distribution optimal parameter estimation method, comprising the following steps: S1, for each parameter to be estimated within the scope of a domain, respectively utilizing a variable grouping method and dividing a theoretical distribution of a skewed distribution into a plurality of groups; S2, calculating a statistic Z for each parameter to be estimated after variable grouping with respect to the theoretical distribution, and selecting the parameter to be estimated corresponding to the statistic with the smallest numerical value to serve as an optimal parameter estimation value. The present invention can obtain an accurate estimation parameter, solving the problem where some forms of skewed distribution have an inestimable parameter, as well as the problem where an imprecise estimate during a model goodness of fit test is often unable to pass. Also, the present invention is suitable for many fields which conform to a skewed distribution model, such as linguistics, life sciences, and computer science, possessing important value for scientific research and practical applications.