Wikipedia says kurtosis is the measure of the "peakedness" of the probability distribution of a real-world random variable. Higher kurtosis means more of the variance is due to infrequent extreme deviations as opposed to frequent modestly-sized deviations. Then, they show this image with the following caption:
"Caption: The far red light has no effect on the average speed of the gravitropic reaction in wheat coleoptiles, but it changes kurtosis from platykurtic to leptokurtic (-0.194 -> 0.055)."Image from Wikipedia, shown under GNU Free Documentation License.
Posts
2 comments:
In more readable English, kurtosis simply measures how heavy the tails of your bell curve is. If the tails have lots of area under them (1st pic), then kurtosis is large. If the tails don't have much area (2nd pic), kurtosis is small.
hahaha, thanks Eric!
Post a Comment