In the fields of probability and stochastic processes, the term reasoning refers to the ability to arrive at a conclusion in the presence of incomplete and, potentially, noisy observation. Noisy observation, in turn, refers to an observation (measurement) in which the fact (truth) is mixed with an uncorrelated and erroneous input that may stem from the source of the observation as well as the channel.
In general, we apply three established techniques to reason about an observation. They are called smoothing, filtering (estimation), and prediction. Which of these techniques are suitable for a given observation depends on whether you want to reason about the past, the present, or the future. If it is about the present, the technique we apply is called filtering (estimation).
If we apply some of the fundamental techniques of smoothing and estimation to reason about the Prime Minster of Ethiopia, we will conclude with reasonable confidence that he was pronounced dead on Sunday, July 15, 2012 or even before that. The most probable time, however, is Sunday, July 15, 2012, between 4:00 and 5:00 AM. Likewise, the location of his death would be Saint-Luc Hospital in Brussels, Belgium.
Some of the facts pertaining to the observations I used in my analysis are summarised as follows:
The sources which delivered the news regarding the illness of the Prime Minster and the hospital in which he received treatment proved themselves remarkably reliable. Hence, they should also be given some credibility when we consider their assertion concerning the type of his illness, which is blood cancer.
None of these sources were official sources or sources close to any government (Belgium, Ethiopia, US, British, etc.). In probability and stochastic reasoning, this truth has a significant value.
The government of Ethiopia has so far refused to reveal the type of illness as well as the place where the Prime Minster is allegedly receiving treatment. Apparently, one of the reasons for this is to hinder reasoning; in the presence of this knowledge, one can very easily draw a conclusion.
In fact, no one claims knowledge of the whereabouts of the Prime Minster.
The morning of July 15, 2015 (Sunday) is the richest day in the generation of the observations.
The government of Ethiopia is unable to deliver a proof about the existence of the Prime Minster.
The Wife of the Prime Minster returned to Ethiopia over the weekend (without her husband). Unless his recovery or death was irrefutably convincing to her, it is beyond human reasoning that she would be strong enough to leave him in uncertain condition and be swayed to return to Ethiopia. Looking into the past, his recovery could not have appeared to her certain. Hence, the reasonable assumption is that she was convinced of his death.
The video footage showing the confrontation of journalist Abebe Gellaw with the PM on May 18, 2012, at a Symposium on Global Agriculture and Food Security at the Ronald Reagan Building in Washington, D.C., visibly shows that the PM experienced an untimely intense internal pain and was struggling to moderate his breathe. A close scrutiny of this evidence is revealing. To begin with, even though the Prime Minster received the pain unexpectedly, he did not appear to be altogether surprised by it. People who suffer from a sudden attack (for example, a cardiovascular attack), will usually be unable to temporarily control their breathing, gaze, and gesture. The Prime Minister, on the other hand, took time to nurse the pain and to regulate his inhalation. From this, one can assume that the Prime Minster was suffering a prolonged pain which could easily be aggravated by an emotional stimulus or as sudden release of body chemical.
The relationship between (3) and (9) leads us to another irresistible proposition, namely, the Prime Minster could have been suffering from a gastric tumour called adenocarcinomas (in other words, stomach cancer), possibly, level 3 (5 or more than 5 invaded lymph nodes in the resected specimen).
I assigned to the above propositions some conditional probabilities and employed a Bayesian Network to compute the Likelihood of the death of the Prime Minster. Even by the strictest assignment of probabilities, the outcome strongly indicates that it is highly probable that the Prime Minster of Ethiopia died on Sunday, July 15, 2012 between 04:00 and 05:00 AM. In contrast, the collective conditional probability for the propositions: “he is alive”, “he is recovering”, “he has recovered”, and “he hasn’t been ill” was significantly small.
