prediction

[mark.wilcox@homecall.co.uk]

The debate around the notion of prediction is quite interesting because it raises some generic problems "we" have in coming to terms with the fact that while our 'minds' are quite useful devices, we continue to search for mathematical or mechanical solutions to help us do what we do naturally. That is, making (or trying to make) sense of the present situation in relation to our expereince of the past in order to predict the future. Myron's short but pertinent note alluded to this. The title of Lewis' first book also provides a clue "Mind and the World Order"
  
Didn't Dr Deming say that the control chart was no substitue for the brain?

Prediction is an ongoing process assisted by the control chart. To make sense of the present and predict the future we use what Lewis calls probability-judgements.  Three characteristics associated with probability-judgements reflect the pragmatism in his thesis.  

Lewis’ Probability Characteristics
1	The probability judgement does not have to be true – but practical – and capable of revision.
2	The probability judgement is relative to the pertinent knowledge of the person making it and this does not bar its validity.
3	The probability judgement is relative to the particular circumstances in which it is made, so if it is valid it is true for this particular instance

Probability Characteristics adapted from Lewis, 1929:346 

These characteristics are part of his theory of knowledge, based upon an interpretation, using a person’s cumulative knowledge to assess the particular circumstances in which it is made.  This approach is quite typical of the pragmatic movement at that time, because of it being free from metaphysical dogma, and rooted in the ‘here and now’ pragmatism of the person.  It should be noted that Shewhart uses the notion of ‘judgement’ or ‘human-judgement’ having found the application of probability theory too slow and cumbersome to be of practical use for his work.    
Inevitably prediction is a human process.
We see now how the person, or to be more precise the mind, is central to Lewis’ thesis and we should note the important premise that data do not interpret themselves.  It is a human activity, or process, of arguing from past to the future with the help of probability-judgements.    Lewis helps define this process: 
“the statistical prediction of the future from the past cannot be invalid, because what-ever is future to any given past, is in turn past to some future.  That is, whoever continually revises his judgement of the probability of a statistical generalization by its successively observed verifications and failures, cannot fail to make more successful predictions than if he should disregard the past in his anticipations of the future.  This might be called the ‘Principle of statistical accumulation’.  It is quite evident that it holds even with respect to what is determined ‘pure chance’ – in the only sense in which we conceive anything such.  This is what is meant by saying that probability or chance is measured by that fraction which is approximated ‘in the long run’ (Lewis, 1929:386).  
So clearly arguing from past to future, which is the basis of prediction, is a human activity and relative to experience and the person or person’s mind involved.  The principle of statistical accumulation is an interesting concept that Lewis expands on to some degree to show how even in extreme circumstances we are able to make sense of data in the long run.
  Lewis was quite clear that the mind had an active, critical and discriminatory role in interpreting experience.
An important feature of any prediction is the need to be clear on the purpose of what it is we are trying to achieve.  Without a clear purpose we would struggle to make a valid interpretation.

I think this is how Shewhart interpreted Lewis....
“The successful quality control engineer, like the successful research worker, is not a pure reason machine but instead is a biological unit reacting to and acting upon an ever changing environment” (Shewhart:38).  
This fascinating description highlights the role of the mind in managing an organisation in flux with the use of statistical data and control charts.

Sorry its a long post - but Ive done quite a bit of research on this fascinating topic.

Dr Mark Wilcox
 

Message posting through the Clemson CQI Web Server.