Inferring
and
(
possibly different from
)
After having been playing with
's and their
ratios, from which we started for simplicity, let us
move now to the rates
and
of the two Poisson processes,
i.e. to the case in which the observation times
and
might be different.
But, before doing that, let us spend a few words on the reason
of the word `deducing', appearing
in the title of the previous section.
Let us start framing
what we have been doing in the past section in the graphical model
of Fig.
,
known as a Bayesian network
(the reason for the adjective will be clear in the sequel).
Figure:
Graphical model showing the model underlying
the inference of
and
from the observed
numbers of counts
and
, followed by the deduction of
.
 |
The solid arrows from the nodes
to
the nodes
indicate that the effect
is caused by
,
although in a probabilistic way (more properly,
is
conditioned by
, since, as it is well understood
causality is a tough concept16).
The dashed arrows indicate, instead,
deterministic links (or deterministic
`cause-effect' relations, if you wish).
For this reason we have been talking about `deduction': each couple of values
provides a unique value of
,
equal to
, and any uncertainty about
the
's is `directly propagated' into
uncertainty about
. The same will happen with
.
Navigating back along the solid arrows, that is from
to
, is often called a problem of `inverse probability',
although nowadays
many experts do not like this expression, which however
gives an idea of what is going on. More precisely, it is
an inferential problem, notoriously tackled for the first time
in mathematical terms by Thomas Bayes [19]
and Simon de Laplace, who was indeed talking about
“la probabilité des causes par les événements”[20].
Nowadays the probabilistic tool to perform this `inversion'
goes under the name of Bayes' theorem, or Bayes' rule,
whose essence, in terms
of the possible causes
of the
observed effect
,
is given, besides normalization, by this very simple formula
having indicated again by
the background state of information.
quantifies our belief that
is true taking into account
all available information, except for
.
If also the hypothesis that
is occurred
is considered to be true,17then the `prior'
is turned into the `posterior'
.
We shall came back on the question of the priors,
but now let us move to infer
,
and their ratio
.
Subsections