Models Are Not Reality

There was a time when building model airplanes was a fun thing to do.  It may have been the fumes from the glue, I suppose.  I don’t think we thought that an engineless model F-86 Sabre was real.

f-86But that has changed some.  Now people reverentially refer to models as if they are observed reality.

As discussed a few weeks ago making decisions on wrong or incomplete “facts” leads to disaster.  Minor knowledge of how models build and work is helpful.

  • Models never include all important information.  Some of it is too hard to acquire or is only important some of the time.  Other information is wrong.  Other is over summarized.
  • Models provide no answers.  They are created to provide insight.  Sometimes they will help you discover missing pieces.
  • Models are built on information from the past and if future information will be different, they are likely valueless.
  • Models are built on information that was contextually correct when it was created.  Unless careful attention is paid to present and future contexts, the information may not be relevant.  Sometimes the data can be adjusted for anomalies in context, sometimes not.  When adjusted information forms part of the model, you must know the method for adjustment.
  • Finally, and perhaps most commonly overlooked, models do not create conclusions, they create a way to see relationships amongst the variables.  This observation allows people to understand the limits to their resources and their application in the real world.

For example if I build a model for retirement planning I may choose 5% as a long term average investment yield.  If I assume inflation of 2%, I may be okay, but not with certainty.  It is possible that yield will be inflation plus 3% but difficult.

So the model runs at a 2% spread, 5% and 3% and it shows that savings must be higher or retirement delayed.  Knowing how much the difference is becomes useful.  Running it at 9% and 7% looks like the same 2% spread but taxation is a serious impediment as yields go up.  The model says, “Learn something about tax management.”

Models are would-be worlds.  They are accurate under only two conditions.

  1. All of the assumptions within the model reflect future reality and are correctly described and weighted.
  2. All of the other possible and ignored assumptions are irrelevant and or mutually cancelling.

Good luck with meeting both of those criteria.

The media would have you believe that the future can be known within limits.  It cannot.  The information is too chaotic for easy prediction.  The most we can hope for is a model that identifies the essential problems and identifies their dimensions.  By applying the 3Rs of planning, record, review and revise, we can develop skill and better information derived from what was once the future.

Continuous adjustment is the key.

Without an understanding of how the past and present came to be as they are and what other variations may come to be, there is no strong basis for projecting future outcomes.

Think tendencies (meaning) not results.

Don Shaughnessy is a retired partner in an international public accounting firm and is now with The Protectors Group, a large personal insurance, employee benefits and investment agency in Peterborough Ontario.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: