1
$\begingroup$

In my econometrics and economics classes, the term autocorrelation has come up this quarter. I have read the Wikipedia pages about it and tried to understand them, but they are a bit beyond me. They use too much math from signal analysis that I don't have time to study at the moment. But I would still like to understand the general idea.

Can someone provide a brief explanation of what autocorrelation is and how this concept is used to understand price and analyze price trends?

$\endgroup$
1
  • 1
    $\begingroup$ statisticssolutions.com/autocorrelation That contains intuitive explanations with little or no math. It also includes a few useful examples. Check it out. $\endgroup$
    – 123
    May 23, 2016 at 15:47

1 Answer 1

2
$\begingroup$

Autocorrelated is the opposite of independent, which is a term easier to understand (and explain). If you throw one die, the probability of getting the number any number is 1/6. If you throw it again, the probability of guessing the result number is 1/6. If you do it again and again and again the probability of getting it right is always 1/6 (provided that the die is «fair»). This means that the next outcome is independent to the previous one. Whether you get a 3 or a 5, the outcome of the next «experiment» is not affected.

The opposite would be, for instance, if you have a set of 20 cards, and then you draw one, the probability that the card drawn is the one you predicted is 1/20th. But this time, things have changed: you are left with only 19 cards on your hand. So the next time, you have a probability of 1/19th of guessing the card. If you do it again and again the probability of guessing the next card is greater. Since the outcome of the previous event helps to predict the next one, we say that this process is autocorrelated.

One doesn't need to be an expert on statistics to see that these two examples are different. This is why it is important.

Now, when it comes to prices, it simply means that the price of a good today helps predicting its price for tomorrow. Take currencies: If 1€ = USD \$1.1193 today, it would be very unlikely (but never impossible) that 1€ = USD \$10 tomorrow. It is more likely that the price will be around USD \$1.1193.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.