Capability Analysis (Normal Distribution) - Transformmain topic see also
Capability Analysis (Normal Distribution) - TransformStat > Quality Tools > Capability Analysis > Normal > Transform
Allows you to specify how to transform your data.
Choose No transformation when your raw data follow a normal distribution. See Individual Distribution Identification if you are unsure which distribution adequately fits your data.
Choose the Box-Cox or Johnson transformation to transform nonnormal data to follow a normal distribution. The output includes a histogram for the original and transformed data with their specification limits for comparison. You can also store the transformed data for further analysis. If the one transformation does not adequately transform your data, the other may work better.
For the Box-Cox transformation, you may select a method for determining lambda. The common transformations are natural log (l = 0) and square root (l = 0.5). You can also choose any value between - 5 and 5 for l. In most cases, you should not choose a l outside the range of -2 and 2. You may want to first run the command Stat > Control Charts > Box-Cox Transformation to help you find the optimal transformation value. You can only use the Box-Cox transformation with nonnegative data. Minitab uses rounded values for Box-Cox transformations when possible. If do not want to use rounded values for l, see Tools > Options > Control Charts and Quality Tools > Other.
For the Johnson transformation, you may enter the p-value for selecting the best transformation. Minitab optimally selects a function from three families of distributions of a variable, which are easily transformed into a standard normal distribution. These distributions are labeled as SB, SL, and SU, where B, L, and U refer to the variable being bounded, lognormal, and unbounded. Minitab is unable to estimate the within subgroup variation when using the Johnson transformation.
No transformation: Choose when your raw data follow a normal distribution.
Box-Cox power transformation (W=Y**Lambda): Choose to perform a Box-Cox transformation.
Use optimal lambda: Choose if you want Minitab to estimate an optimal value of lambda.
Lambda = 0 (ln): Choose if you want Minitab to use lambda equal to zero.
Lambda = 0.5 (square root): Choose to use the square root of the data.
Other (enter a value between -5 and 5): Choose and enter your own value of lambda to bypass the lambda estimation step.
Johnson transformation (for overall analysis only): Choose to perform a Johnson transformation.
P-Value to select best fit: Type a p-value for selecting the best transformation.