Definition:Sheppard's Correction
Jump to navigation
Jump to search
Definition
Let $S$ be a set of grouped data.
Let it be assumed that $S$ obeys a Gaussian (normal) distribution.
When calculating the moments of $S$ using the mid-interval values, it is in general necessary to apply an adjustment to the calculated values.
This is called a Sheppard's correction, and it is subtracted from the calculated moment thus:
\(\ds \hat \mu_2\) | \(=\) | \(\ds m_2 - \dfrac {h^2} {12}\) | ||||||||||||
\(\ds \hat \mu_3\) | \(=\) | \(\ds m_3\) | ||||||||||||
\(\ds \hat \mu_4\) | \(=\) | \(\ds m_4 - \dfrac {m_2} 2 + \dfrac {7 h^2} {240}\) |
where:
- $m_i$ is the $i$th moment calculated from the mid-interval values
- $\hat \mu_i$ is the adjusted value of the $i$th moment
- $h$ is the bin width.
Also see
- Results about Sheppard's corrections can be found here.
Source of Name
This entry was named for William Fleetwood Sheppard.
Sources
- 1998: David Nelson: The Penguin Dictionary of Mathematics (2nd ed.) ... (previous) ... (next): grouped data
- 2008: David Nelson: The Penguin Dictionary of Mathematics (4th ed.) ... (previous) ... (next): grouped data
- Weisstein, Eric W. "Sheppard's Correction." From MathWorld--A Wolfram Web Resource. https://mathworld.wolfram.com/SheppardsCorrection.html