"Lies, damned lies, statistics ! ” How this may influence your decision making in Investing ?
How to Lie with statistics!
As an investor, we have been bombarded by news and statistics every minutes and second, as we know well, some of this news been exaggerated and statistics or figures used also questionable.
We need to understand the figure behind these statistics and how is it derived to ensure that we have not been lied by statistics.
Darrell Huff gives people the tools to talk back to statistics.
"There is terror in numbers and nowhere does this terror translate to blind acceptance of authority more than in the slippery world of averages, correlations, graphs, and trends. “ writes Darrell Huff.
What was true in 1954 is just as true today. According to Huff, below are just a few common tactics used to influence readers by using some “twisted or distorted “ ways of presenting their data in the presentation.
“The Gee-Whiz Graph" and “ The One Dimension Picture “
This is about how the graphical display of statistics can be twisted so that one can get any desired result, though the statistics figure aren't changed. Such graphic or picture will give us “ eye illusion “ in order for the presenter to achieve their desire message or result.
For example, the below charts show the effect of “impact “ by changing the scale and starting point of the Y-axis, by doing this, the presenter would be able to “shock the reader “ and win the argument.
Correlation doesn’t mean causation :
Duff also points out the fallacy of correlation. Sometimes the two variables have a very strong correlation and both the statistical and mathematical requirements are met but that doesn’t mean it has causation.
For example, there is a strong correlation between a school child's height and the child's score on a given spelling test - taller kids do better. The fact is a lot less surprising when you see that first graders tend to be smaller than sixth graders, and tend to know fewer words. Maybe the example sounds silly but no sillier than lots of the numbers in the news every day.
'Skirt Length Theory' in Wall-street.
|<Image credit to: socionomics.net>|
The idea that skirt lengths are a predictor of the stock market direction. According to the theory, if skirts are short, it means the markets are going up.
And if the skirt is long, it means the markets are heading down. Also called the Hemline Theory.
What do you think?
Below weblink shown some of the very fascinating charts in plotting two different variables in very high correlation but no causation at all.
Few charts to shown from the above link: just for fun!
<image credit : fastcodesign.com>
Poorly chosen and the problem of average :
“Statistician, a person who lays with his head in an oven and his feet in a deep freeze stating, On the average, I feel comfortable”- Bruce Grossman.
Below are the annual returns for the S&P 500. The red circles represent the years that were within half a per cent of the average return. If the average return has occurred in just three out of ninety years, you should probably be very sceptical of anyone crafting a narrative based solely on historical averages.
Same apply to STI Index investing, when one buying into Index ETF base on long term average return of 7-8%, but bear in mind that it was based on a long period of average, at anyone time, the return could be from +/- 20 to 40%. Investors may feel “shock” that even with ETF Index investing, the volatility could be huge on a yearly basis.
This is another one of my favourite charts on “statistically bias on average “.
The sample with Build-in bias.
Quoted from Wikipedia: “In statistics, sampling bias is a bias in which a sample is collected in such a way that some members of the intended population are less likely to be included than others.
It results in a biased sample, a non-random sample of a population (or non-human factors) in which all individuals, or instances, were not equally likely to have been selected. If this is not accounted for, results can be erroneously attributed to the phenomenon under study rather than to the method of sampling.”
The most common type of sampling bias is “ Selection from a specific real area.” For example, a survey of high school students to measure teenage use of illegal drugs will be a biased sample because it does not include home-schooled students or dropouts.
A sample is also biased if certain members are underrepresented or overrepresented relative to others in the population.
For example, a "man on the street" interview which selects people who walk by a certain location is going to have an overrepresentation of healthy individuals who are more likely to be out of the home than individuals with a chronic illness. This may be an extreme form of biased sampling, because certain members of the population are totally excluded from the sample.
Ways to Avoid Being Fooled By Statistics
This LINK may give you some clues :
I would like to end my post by below link of Video from TED-Talk: by Sebastian Wernicke on Lies, damned lies, statistics.
Quote Of The Day :