Modelling Data Containing Exact Zeroes Using Zero Degrees of Freedom
Andrew F. Siegel
Abstract
Andrew F. Siegel
Abstract
SUMMARY The noncentral chi-squared distribution with zero degrees of freedom can be used to model continuous data of a variety of shapes that also contain exact zero values. Existence and uniqueness of the maximum likelihood estimates for the scaled non-central chi-squared distribution with zero degrees of freedom are established, computational methods are considered, and an illustrative example is given.
OpenAlex reports 32 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
SUMMARY The noncentral chi-squared distribution with zero degrees of freedom can be used to model continuous data of a variety of shapes that also contain exact zero values. Existence and uniqueness of the maximum likelihood estimates for the scaled non-central chi-squared distribution with zero degrees of freedom are established, computational methods are considered, and an illustrative example is given.
Key concepts: Degrees of freedom (physics and chemistry), Zero (linguistics), Mathematics, Uniqueness, Distribution (mathematics), Applied mathematics, Square (algebra), Mathematical analysis