Skip to content

Questioning Assumptions and the Wizard of Oz

June 25, 2010

For those that follow the blog, it would probably not surprise you that I have never been one to  take old theories or new concepts at face value. The root of truth comes from asking good questions that get to the root of the assumptions made by a given theory or research finding–no matter how elegant. The sad fact is that in the world of economics and finance, tackling a journal article involves making sense of  a sea of incomprehensible numerical symbols and mathematical proofs. When you combine that with dense tables and strange acronyms for arcane statistical procedures, it is no wonder that most give up after the first few pages. Most authors could qualify to be math professors at any university, and could readily solve the area of the pyramid without a calculator. The veil of numerical and statistical rigor is just as intimidating as the Wizard of Oz. The comparison couldn’t be more appropriate, because often as you dig deeper into the underlying logic of many of these articles the voice beckons: “Pay no attention to the man behind the curtain!”

If you can truly keep yourself awake and focus on each line of math, and each sentence of methodology—-the premises and assumptions are laid bare for the skilled prosecutor to attack. If you fail to pay attention, the message  of the paper seems unquestionable and you are left feeling helpless but to accept its conclusions. But this effect is not specific to finance or economic papers—no, this effect is found in almost all the news articles and financial media. It is also found within the blogosphere (myself included) where substantiation is secondary to content.  Many of the gurus especially on TV are guilty of speaking way over people’s heads, and quoting facts that no one could possibly verify which helps them in making conclusions that no one can question. Put on a suit and tie, some grey hairs and a respectable appearance and it becomes even harder to question their version of the truth.

Whenever you listen to an argument, opinion, or  theory, ask yourself this: 1) what are the premises of the argument that are being made? 2) what are the hidden premises that are implied but not explicitly mentioned 3) what are the facts/evidence being used to substantiate a given conclusion etc 4) do the facts and evidence logically support the conclusion 5) is/are the facts and evidence being measured possibly spurious or inaccurate? could they possibly be interpreted in a different way—if so what might they also imply? You might think that this process is a waste of time—or be the type of person who is focused on backtests, the latest Goldman report,  and the performance metric du jour. If so you are totally missing the point—the numbers cannot and most certainly will not tell you about the things that you haven’t considered in designing a system for example. The facts that you have in front of you cannot tell you what you haven’t thought about when you make a qualitative decision to buy a stock or commodity.  Most of us are victims of the need to seek confirmation and validation due to our inherent insecurity about making decisions under conditions of uncertainty. As a result whatever you have to support your decisions– whether advice from a guru or your own research or backtesting— will always seem more thorough and accurate than it actually is. Its up to you to find out which is which.

One Comment leave one →
  1. June 25, 2010 1:01 pm

    David, best one yet! It’s resonating in the wind: “Do you really think the emperor cannot be naked in this Day and Age?”🙂

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: