Subsequently, a cautionary response was posted by Michael Lamoureux over at Sourcing Innovation [2]. In his post, Lamoureux emphasizes that a SRA model is useful for generating a point estimate (baseline) but constrained by the inherent assumptions involved in performing regression analysis. More specifically, Lamoureux states that the model would not be applicable if “a new buyer comes in that totally redefines the demand and the market strategy, or the market conditions have suddenly changed from supply shortage to supply surplus, or new production technologies could revolutionize production and trim overhead 20%.”
Read in conjunction, these two articles afford an
exceedingly important point: Predictive analytics is very powerful, and
should be employed frequently by procurement professionals, as long this is done in
an intelligent and responsible manner. I cannot emphasize enough how
important the words “intelligent” and “responsible” are in that sentence. In
general, mathematical models can be described as simplifications of the real world
in some that we may construct a computationally tractable understanding of it. As
such, they have limitations. It is the responsibility of both managers
and analysts to understand what these limitations are.
In order to put Lamoureux’s statement on a more concrete
foundation I would like to explore a moment in history during which some very intelligent people forgot to heed his advice. Recall Black Monday
(October 19, 1987). Leading up to this day, the financial market had become dominated
by quants. The Black-Scholes equation, sometimes referred to as Midas Formula since
it was believed to be a recipe for making everything turn to gold, was in widespread
use for pricing options. Options are one of the oldest and simplest derivatives; derivatives are essentially bets upon bets. The development and widespread
adoption of options led to ever-more complicated derivatives and just as many
complicated mathematical models attempting to underpin their true value.
Rather than delve into financial literature, I will attempt
to explain how a derivative might work through an analogy. Suppose that you bet
your friend John $100 on the outcome of a football game. Say the Browns versus
the Patriots. You choose the Patriots of course. At the end of the first
quarter the Patriots are up 14 to 3. Rather than wait to see if the Patriots do
in fact win, you would prefer to walk away with a guaranteed profit. You sell the
right to the $100 bet to our friend Bill. Since it is very likely that the
Patriots will win if the score is 14 to 3 already, Bill says that he is willing
to pay you $20 for the right to the bet. You have now guaranteed your profit by
creating a derivative (selling the right to your bet) and Bill has now assumed all
the risk (albeit, he has a pretty good chance of seeing a payout for a nominal fee).
While the assumptions of the Black-Scholes model are beyond
the scope of this article, it is widely accepted that when they hold true then
the risk of using it is low [3]. Of the utmost importance to us is that the
model, and other similar models, are only applicable when the market is stable.
On Black Monday, this assumption was violated to an extreme. The world’s stock
markets saw a loss of more than 20% of their value within a few hours. While
the Black-Scholes equation was not sole the cause for the financial
crash, it was “one ingredient in a rich stew of financial irresponsibility,
political ineptitude, perverse incentives and lax regulation” and “may have
contributed to the crash … because it was abused.” [4]
Post A Comment:
0 comments so far,add yours