Causation and correlation 101

It appears there a gaping hole of disconnect between causation and correlation.  Correlation says that two things move along in equal or almost equal ways or opposite ways.  Causation says that one thing affects the movement of another. Huge difference.

Lets talk about two recent examples:

The market has been up X% on Tuesday’s after X event occurred.

The incumbent has won 17/18 times when the Redskins have won the game before the general election.

In understanding the difference between causation and correlation you must first make a judgement  if they are connected in anyway at all.

In the first example, it could be true that something in the market is happening that causes it to move in a certain way.  I will get into this more in the bottom.

The second example, it would be hard to keep a straight face if someone argued that the Redskins winning affected the outcome of a general election.

How you can use the information:

In the second example, it is a cool stat and will probably get you a bunch of retweets but it is info porn.  The first example could be useful.  Things to consider:

  • Does it happen enough times to dig through the data and find out what is behind the number?  We do not have unlimited time and attention spans.
  • Is the event most important to market participants?  Will participants act in a predictable given this information or do they even really care?
  • Is the event in the same context as past number?  If you change the equation the answer changes or at least the “meaning” of the answer.
  • Where is the information heading?  Is it too late, are you going to be the greater fool?
To borrow and modify a line  from JC Parets, we do not have data from the future.  Statistics record they do not tell a story.  Probabilities force you to look into the story but the context and plot is developed by the person using it.  Correlations vary by closeness and duration.  If you find causation, examine it deeply, but quickly because they often change too.

We are in the processing of information business and the point of the business is to make money. Get rid of the shit you don’t need.  Do you want to be the smartest guy in the room or the one continually getting closer to your goals?

Sign up for our FREE Morning Brief, big picture view of markets in critical area and proprietary support and resistance. You can also sign up to see us trade live here.

We would really appreciate your feedback, if you like, hate, or think we are full of crap. Please leave a comment, a voice mail (312) 725-9121 , email info @ traderhabits (dot) com or twitterstocktwitsyoutube and facebookSubscribe to Traderhabits by email or to newsletter.

  • jim nazyum

    no, i am sorry.

    the redskins lost and that means obama is going to lose.

    don’t fight the trend.

One Response to Causation and correlation 101

  1. jim nazyum says:

    no, i am sorry.

    the redskins lost and that means obama is going to lose.

    don’t fight the trend.

The services and materials provided should not to be interpreted as investment advice, an endorsement of any security, commodity, future, or personal investment advice, or an offer to buy, sell, hold or trade futures, options or commodity interests or a recommendation to buy, sell, hold or trade futures, options or commodity interests. You assume the entire cost and risk of investing and are solely responsible for any and all gains and losses, financial, emotional, or otherwise, experienced, suffered, or incurred by you.

Read previous post:
Looking before you cross.

I do not know what it is lately but it seems like I see a person risk their life or...

Close