Im trying to undestand this qqplot from arima residuals but im a bit lost about the underlying distribution, concretely I don't now how to interpret the tails.
$\begingroup$
$\endgroup$
3
-
$\begingroup$ Basically you have some long tails on your distributions for your residuals. You may want to inspect what data (observations) may be causing this. It may be just some outliers or something more serious that you haven't taken into account in your ARIMA model. $\endgroup$– JonCommented May 17, 2019 at 22:59
-
$\begingroup$ Here is a nice reference on how to interpret QQ plots: seankross.com/2016/02/29/A-Q-Q-Plot-Dissection-Kit.html $\endgroup$– JonCommented May 17, 2019 at 23:02
-
1$\begingroup$ There's several relevant posts right here in site about interpreting Q-Q plots, such as How to interpret a Q-Q plot $\endgroup$– Glen_bCommented May 18, 2019 at 0:02
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
1
My interpretation would be that the middle values of the sample are close to what you'd expect from normally distributed data, as it follows the straight line from the diagram closely.
However, it seems the underlying data distribution presents extreme values more often than a normal one, that's why you see the points going under the line for big negative values and over it for big positive ones.
If you want to check that statement, you may want to get the sample kurtosis (see https://en.wikipedia.org/wiki/Kurtosis ) Data from the finantial world often comes with high kurtosis. Are you working on asset returns or something related?
-
$\begingroup$ David my data is about some physical phenomena, concretely distinct types of temperatures in the short term. And yes it has extreme values at day and night, my forecasting its just about this points, but im failing in capturing this behavior. $\endgroup$– VYagoCommented May 18, 2019 at 0:44