Blog 41: Correlation and Causation by Xavier Lo, FIA, FRM, MBA

I think most of you would have heard of correlation [相關性] versus causation [因果關係]. The idea here is that just because two things are correlated (in other words, when one goes up, the other goes up by the same proportion [比例]) doesn’t mean that one causes the other.

Here’s a good example. If you look at the number of swimming accidents [游泳意外] and the sales of ice cream [雪糕銷售量], you’ll see that they follow the same patterns: probably highest during the Summer months, and lowest during Winter. They are definitely correlated, but does this mean that eating ice cream makes you a really bad swimmer? Potentially, but the likelier explanation is that there’s a another factor [另一個因素] that causes both of them – the temperature [溫度]! Now you can see that when the temperature rises, more people go swimming (and hence more accidents) and more people eat ice cream.

Although you might think the example makes it obvious the differences between correlation and causation, it might not be so easy separate this during work projects. Its obvious that ice cream sales and swimming accidents are correlated but have no causal link, but how about car accidents [車禍] and the colour of the car [車顏色]? Is there a third factor in play (like the temperature in our example), or is the car colour really something that causes more accidents (maybe a black car is not easy to see at night)? The answer isn’t obvious. Also, when we are so concentrated in our data crunching and are really pressed to make an amazing discovery, it is so easy for us to suddenly force an explanation [強行解釋] with the patterns that we see.

What separates us from robots is that we have common sense. Whenever we see two correlated factors, really stop and ask yourself: is this causation, or just two correlated things that are driven by some explanation that you haven’t seen yet? Don’t fall into the trap of thinking everything is causal!

About the Author

Xavier Lo, FIA, FRM, MBA

Qualified fellow actuary (in UK and Hong Kong), Financial Risk Manager, and MBA graduate (listed on the Dean's List) with a passion for insurance, data science, and analytics. Experienced in a broad range of insurance roles (pricing, capital modelling, reserving, ERM), along with a touch of knowledge in banking. Member of the General Insurance Committee (2021), Actuarial Innovation Committee (2019 - 2021) in ASHK.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.