Subscribe to DSC Newsletter

Although this is what my Army platoon Sergeant often told me—he would, of course, precede it with the proper address of “Sir”—it really makes a lot of sense; however, I did not realize that my platoon sergeant was such a student of humanity, until much later when I heard about this fellow named Occam.

William of Occam, or sometimes rendered Ockham, (c. 1287–1347) is remembered as an influential medieval philosopher and nominalist, though his popular fame as a great logician rests chiefly on the axiom attributed to him and known as lex parsimoniae (Latin: literally, “the law of economy”) or Occam’s razor. The term razor refers to distinguishing between two hypotheses either by “shaving away” unnecessary assumptions or cutting apart two similar conclusions.

This axiom seems to represent the general tendency of Occam’s philosophy, but it has not been found in any of his writings. [1] His nearest pronouncement seems to be “Numquam ponenda est pluralitas sine necessitate” [Plurality must never be posited without necessity], which occurs in his theological work on the “Sentences of Peter Lombard” (a commentary known as Ordinatio) [2].

The words attributed to Occam, “Entia non sunt multiplicanda praeter necessitate” (Latin: “Entities must not be multiplied beyond necessity”), are absent in his extant works; this particular phrasing owes more to John Punch. [3] Indeed, Occam’s contribution seems to be to restrict the operation of this principle in matters pertaining to miracles and God’s power: so, in the Eucharist, a plurality of miracles is possible, simply because it pleases God. [4]

This principle is sometimes phrased as “Pluralitas non est ponenda sine necessitate” (Latin: “plurality should not be posited without necessity”). [4] In his Summa Totius Logicae, (Latin: “Sum of Logic”) i. 12, Occam cites the principle of economy, “Frustra fit per plura quod potest fieri per pauciora” [Latin: “It is futile to do with more things that which can be done with fewer”]. [5][6]

Einstein himself expressed caution when he formulated Einstein’s Constraint: “It can scarcely be denied that the supreme goal of all theory is to make the irreducible basic elements as simple and as few as possible without having to surrender the adequate representation of a single datum of experience”. An often-quoted version of this constraint (that cannot be verified as being posited by Einstein himself) [7] says, “Everything should be kept as simple as possible, but no simpler.”

When we apply it in Operations Research, Data Science, Predictive Modeling and Analytics, it simply means this: when faced with two alternative models, solutions, explanations, etc. with similar or equal predictive power, choose the simplest. The real world that we attempt to mimic is wrought with phenomenon that we cannot capture, possesses variables that we cannot fathom, and often hides explanation of itself within it chaotic behavior (see my post on Chaos theory, “It’s a Jungle Out There”, September 27, 2014), so we make simplifying assumptions in order to construct mathematical or statistical models. This led George Box to say, “All models are wrong; some are useful”. Therefore, if we can explain a consumer response to a marketing campaign, or the probable location of improvised explosive devices (IEDs), just as well with 10 variables than with 20, adopt the simplest.

References

1. Vallee, J. (2013). “What Ockham really said”. Mon, Feb 11, 2013. http://boingboing.net/2013/02/11/what-ockham-really-said.html. Retrieved 2014-10-5.

2. Guillelmi de Ockham. Scriptum in Librum Primum Sententiarum, Ordinatio, Opera Theologica (Latin). Published May 1, 1977 by Franciscan Inst Pubs. ISBN-13: 978-1576590171.

3. Alistair Cameron Crombie (1959), Medieval and Early Modern Philosophy, Cambridge, MA: Harvard, Vol. 2, p. 30.

4. James Franklin (2001). The Science of Conjecture: Evidence and Probability before Pascal. The Johns Hopkins University Press. Chap 9. p. 241.

5. Thorburn, W. M. (1918). "The Myth of Occam's Razor". Mind 27 (107): 345–353. doi:10.1093/mind/XXVII.3.345

6. Spade. P.V. (1995). William of Ockham, From His Summa of Logic – A partial translation of his works. http://www.pvspade.com/Logic/docs/ockham.pdf. Retrieved 2014-10-05.

7. Strickland, J.S. (2011). Albert Einstein: “Nobody expected me to lay golden eggs”. Lulu.com. ISBN 978-1-257-86014-2.

Views: 1261

Comment

You need to be a member of AnalyticBridge to add comments!

Join AnalyticBridge

Comment by Adrian Walker on July 3, 2015 at 8:52am

Here's a simple way of making (big) data useful.

www.reengineeringllc.com/internet_business_logic_in_a_nutshell.pdf

Comment by Chandrasekhara S. "C.S." Ganti on December 3, 2014 at 5:11am

An excellent treatise based on the highest Philosophical comment on the current state of affairs - et tu Brute -- folks joining the fray. Yes, brevity is the soul of wit . and while this is Big Data Science, it does not hurt to explain away with fewer variables, we know we have Billion records .. Religion,  Race, Sex , Income / Savings / Area of Residence  et al  Available potential credit / debt ??? WTP - Willingness to Pay  in case of Consumer Economics / Commercial Products (so half-dozen  / dozen would more than suffice. Not sure .. Anybody who quote the parsimony concepts in selection of  variables from late Prof. Box has got my ear.. " Keep it Simple Not Stupid "-- was always my repartee to the general philosophy. 

On Data Science Central

© 2019   AnalyticBridge.com is a subsidiary and dedicated channel of Data Science Central LLC   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service