Message-ID: <CA+RJqXXPMqudfAXAtbXFpC9YVG76f4UrOX6-=ccMnLcC60DDfQ@mail.gmail.com>
Date: 2016-02-29T10:11:36Z
From: Marco Scutari
Subject: bnlearn and TAN network
In-Reply-To: <000301d172d6$c0f0f640$42d2e2c0$@ecogeonomix.com>
Hi Ross,
On 29 February 2016 at 09:51, <ross.chapman at ecogeonomix.com> wrote:
> The output node is classFFB. My understanding is that the model only
> has one node linked directly to the output node:
> ][matureTreeData.ESTATE|classFFB].
Actually not, e.g.
[matureTreeData.YEAR|classFFB:matureTreeData.ESTATE] means
P(matureTreeData.YEAR | classFFB, matureTreeData.ESTATE). So all
variables depend on classFB.
> I was looking for a plot which shows the learned network, an example of
> which I have attached.
That is the best you can get from bnlearn at the moment - although it
is possible in theory to produce a graph + barplots plot from
Rgraphviz. I have never managed to make it work, though.
Cheers,
Marco
--
Marco Scutari, Ph.D.
Lecturer in Statistics, Department of Statistics
University of Oxford, United Kingdom