Dina [[probability theory]] jeung [[statistik]], '''skewness''' nyaeta ukuran kateu-simetrian [[probability distribution]] tina nilai-[[real number|real]] [[random variable]]. Sacara kasar bisa disebutkeun, sebaran mibanda ''skew'' positip lamun nilai panjang ''tail'' positip panjang sarta skew negatip lamun nilai panjang ''tail'' negatip panjang.
Skewness, [[standardized moment]] nu katilu, diartikeun ku μ<sub>3</sub> / σ<sup>3</sup>, numana μ<sub>3</sub> ngarupakeun [[moment about the mean]] nu katilu sarta σ ngarupakeun [[simpangan baku]]. The skewness''Skewness'' tina random variable ''X'' kadangkala ngalambangkeun ''Skew''[''X''].
ForKeur anilai sample ofsampel ''N'' values the samplesampel ''skewness'' isnyaeta Σ<sub>''i''</sub>(''x''<sub>''i''</sub> − μ)<sup>3</sup> / ''N''σ<sup>3</sup>, where ''x''<sub>''i''</sub> isngarupakeun thenilai ''i''<sup>th</sup> value andjeung μ is thengarupakeun [[mean]].
IfLamun ''Y'' isnyaeta the sum ofjumlah ''n'' [[statistical independence|independent]] randomvariabel variablesrandom, alldina withdistribusi thenu samesarua distribution assalaku ''X'', thensaterusna itditempokeun can be shown thatyen ''Skew'' [''Y''] = Skew[''X''] / √''n''.
GivenSampel samplesnu fromasalna atina populationpopulasi, thepersamaan equationkeur for populationpopulasi ''skewness'' above is anyaeta [[biased estimator]] oftina the populationpopulasi ''skewness''. An [[unbiasedUnbiased estimator]] of ''skewness'' nyaeta is
:<math> \mbox{Skew} = \frac{n}{(n-1)(n-2)}
Baris ka-13:
</math>
wherenumana σ is thengarupakeun sample standardsimpangan deviationbaku andsarta μ isngarupakeun the samplesampel mean.
