moneyFYI
Financial Freedom Is Merely Organized Common Sense
A Statistic Is Not A Fact. It is an Inference
Statistics are frequently necessary and often misleading
Facts are stubborn things, but statistics are more pliable. Mark Twain
There are lies, damned lies, and statistics. Benjamin Disraeli
Statistics are no substitute for judgement. Henry Clay
Math is a language that you use to describe statistics, but really it’s about collecting information and putting it in an order that makes sense. Lauren Stamile
Statistical information is important and can help us make better decisions, but you should not assume you understand. Sometimes people try to trick you and they do so in compelling ways.
How can we learn to use statistical information to our better advantage?
It is like investing.
You won’t go far wrong investing if you don’t make any big mistakes. So too with statistics. If you can use them well, and recognize data or analysis that lacks rigor, you are well on the way.
How do we do that?
Best solutions often just mean understanding what you see. People know what you will find easy and use that against you.
Common statistical tricks:
End point bias. A very common one. Suppose I told you that on 21 September 2000 gold closed at $US 270.20. 21 September 2017 it closed at $US 1,291.50. About 10% rate of return. But suppose I then told you that on 21 September 2012, gold was $US 1,771.23, for an average loss annually of 6.2% annually over the last five years. If I wanted to sell gold I would pick the 10% annual gain. If I wanted to sell something else and you like gold, I would pick the 6.2% loss.
Survivor bias Many populations, the whole field of data, are limited to specific factors. Suppose I told you all fund managers with a 10-year record earned 7% on average. What does that tell you about the manager group as a whole. Nothing whatever. Notice that managers with a 10-year history are kind of rare. Many fail before then, many more started since then. Some have moved firms or changed their fundamental approach. Even if you get the 10-year crowd together, it will say nothing about how they will do next year.
Averages as a descriptor. We are intuitive users of normal distribution. (bell curve) Not all distributions look like that. Some are very peaked in the middle and have bumps on the end, like daily stock returns, while others are U-Shaped. People either love it or hate it.
If the average is presented as meaningful, you can be mislead. There are other measures you might like to know about. Standard deviation is useful sometimes but harder to understand. Mean is average. Median is the amount where half are above and half below. Mode is the most common finding. When all three are the same, you probably have a good bell curve. If the outcome count for average is very low, you should worry. The average Canadian has 1.98 legs, yet there are few or none that demonstrate the characteristic.
Carefully selected populations. If I ask welfare recipients if the monthly amount should be increased, I will get a different answer than if I ask everyone else. If I ask people who are between 20 and 25 if they like Bernie Sanders as president, I get something more positive than from when I ask 70-year-old conservatives. If you don’t know how the population was decided and how the questions were asked, you know nothing.
Presentation bias In Canada the question of Quebec’s succession comes up periodically. If the question is worded in a convoluted way or if it is presented in French alone, the answer will be different.
Know what you can about the population, the selection method, the analysis method and and any smoothing or other manipulation. Then you have a chance of applying common sense and moving forward.
Statistics don’t prove anything about the future.
They are for insight. Statistical inference is the idea to notice. Inference is not a certainty and if the statistician wants to fool you, they can. Statistical data is one of those areas where you can likely get any answer you want if you torture the data long enough. Be a little cautious.
Don Shaughnessy arranges life insurance for people who understand the value of a life insured estate. He can be reached at The Protectors Group, a large insurance, employee benefits, and investment agency in Peterborough, Ontario.
In previous careers, he has been a partner in a large international public accounting firm, CEO of a software start-up, a partner in an energy management system importer, and briefly in the restaurant business.
Please be in touch if I can help you. don@moneyfyi.com 866-285-7772
moneyFYI 