Discrete convolution statistic for hypothesis testing

08/31/2020
by   Giulio Prevedello, et al.
0

The question of testing for equality in distribution between two linear models, each consisting of sums of distinct discrete independent random variables with unequal numbers of observations, has emerged from the biological research. In this case, the computation of classical χ^2 statistics, which would not include all observations, results in loss of power, especially when sample sizes are small. Here, as an alternative that uses all data, the nonparametric maximum likelihood estimator for the distribution of sum of discrete and independent random variables, which we call the convolution statistic, is proposed and its limiting normal covariance matrix determined. To challenge null hypotheses about the distribution of this sum, the generalized Wald's method is applied to define a testing statistic whose distribution is asymptotic to a χ^2 with as many degrees of freedom as the rank of such covariance matrix. Rank analysis also reveals a connection with the roots of the probability generating functions associated to the addend variables of the linear models. A simulation study is performed to compare the convolution test with Pearson's χ^2, and to provide usage guidelines.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/20/2023

Testing distributional equality for functional random variables

In this article, we present a nonparametric method for the general two-s...
research
08/16/2021

Detecting changes in covariance via random matrix theory

A novel method is proposed for detecting changes in the covariance struc...
research
06/18/2023

Optimal test statistic under normality assumption

The idea of an optimal test statistic in the context of simultaneous hyp...
research
09/11/2018

T-statistic for Autoregressive process

In this paper, we discuss the distribution of the t-statistic under the ...
research
03/08/2022

Asymptotic normality in linear regression with approximately sparse structure

In this paper we study the asymptotic normality in high-dimensional line...
research
03/11/2020

A faster and more accurate algorithm for calculating population genetics statistics requiring sums of Stirling numbers of the first kind

Stirling numbers of the first kind are used in the derivation of several...
research
09/08/2019

A hypothesis-testing perspective on the G-normal distribution theory

The G-normal distribution was introduced by Peng [2007] as the limiting ...

Please sign up or login with your details

Forgot password? Click here to reset