glmLink

This function provides characteristics of common link functions (logit, probit, and comlementary log-log). Specifically, based on the link name, the function with its inverse, first derivative, and second derivative is provided.

Provides an expectation-maximization (EM) algorithm using the approach introduced in Warasi (2023) <doi:10.1080/03610918.2021.2009867>. The EM algorithm can be used to estimate the prevalence (overall proportion) of a disease and to estimate a binary regression model from among the class of generalized linear models based on group testing data. The estimation framework we consider offers a flexible and general approach; i.e., its application is not limited to any specific group testing protocol. Consequently, the EM algorithm can model data arising from simple pooling as well as advanced pooling such as hierarchical testing, array testing, and quality control pooling. Also, provided are functions that can be used to conduct the Wald tests described in Buse (1982) <doi:10.1080/00031305.1982.10482817> and to simulate the group testing data described in Kim et al. (2007) <doi:10.1111/j.1541-0420.2007.00817.x>. We offer a function to compute relative efficiency measures, which can be used to optimize the maximum likelihood estimator of disease prevalence.

Md Warasi

groupTesting

Simulating and Modeling Group (Pooled) Testing Data

glmLink function

<dl><dt>fn.name</dt>
<dd>One of the three: "logit", "probit", and "cloglog".</dd></dl>

Arguments

Link Functions in the Class of Generalized Linear Models — glmLink

<dl>

<dt>fn.name</dt>
<dd>One of the three: "logit", "probit", and "cloglog".</dd>

</dl>

glmLink: Link Functions in the Class of Generalized Linear Models

Description

Usage

Value

Arguments

Examples