grad

For the <code>glinv</code> class, which is a high-level user interface, please see <code>grad.glinv</code>; and
for <code>glinv_gauss</code>, which is a lower-level facility, please see <code>grad.glinv_gauss</code>.

A framework for analytically computing the asymptotic confidence intervals and maximum-likelihood estimates of a class of continuous-time Gaussian branching processes defined by Mitov V, Bartoszek K, Asimomitis G, Stadler T (2019) <doi:10.1016/j.tpb.2019.11.005>. The class of model includes the widely used Ornstein-Uhlenbeck and Brownian motion branching processes. The framework is designed to be flexible enough so that the users can easily specify their own sub-models, or re-parameterizations, and obtain the maximum-likelihood estimates and confidence intervals of their own custom models.

Hao Chi Kiang

glinvci

Phylogenetic Comparative Methods with Uncertainty Estimates

Woodrow Kiang

grad function

<dl><dt>mod</dt>
<dd>An object of either <code>glinv</code> or <code>glinv_gauss</code> class.</dd>
<dt>...</dt>
<dd>Further arguments to be passed to the S3 methods.</dd></dl>

Arguments

Compute the log-likelihood gradients of GLInv models — grad

<dl>

<dt>mod</dt>
<dd>An object of either <code>glinv</code> or <code>glinv_gauss</code> class.</dd>


<dt>...</dt>
<dd>Further arguments to be passed to the S3 methods.</dd>

</dl>

grad: Compute the log-likelihood gradients of GLInv models

Description

Usage

Value

Arguments