Home >> Articles >> Common Terms in Yield >> Statistical Terms
Statistical Terms PDF Print E-mail
Written by Aparna Joshi   

Correlation Coefficient

It is an index of the goodness of fit of a straight line to a set of data. A positive correlation value indicates that the two values move in the same direction. Negative value means that the two values move in opposite directions. Zero value shows no linear relationship between the two variables

Cp (Process Capability Potential)

It is the maximum value of CpK given that the observed distribution is Normal with the mean centered between the lower and upper limits. Cp is a difference of Upper Limit and Lower Limit divided by 6-σ of the process distribution. Cp = CpK = UPL - LPL / 6-σ

CpK (Actual or Observed Process Capability)

CpK gives a numerical representation of both the width and alignment of a distribution with respect to the processing limits. CpK is the absolute value of the difference between the process mean and the nearest processing limit divided by 3-σ. If the distribution is perfectly centered within the specified limits, then CpK will be the same as Cp.

CpL = (X-bar - LSL) / 3-σ

CpU = (USL - X-bar) / 3-σ

CpK = Minimum (CpL, CpU)

Chi Square Distribution

This distribution is used to assess the goodness of fit of a hypothetical distribution. For example, the goodness of fit of the Normal distribution to actual data. This test can also be used to see if the distribution is random or not.

For more information on statistical terms, visit http://www.itl.nist.gov/div898/handbook/

More is coming soon... Keep checking...


Joomla Templates by Joomlashack