Hotelling T-Square Control Chart (HTS)

Creates Hotelling T-Square Control Charts

General

Element Name Description
Detail of computed results reported Specifies the level of computed results reported. By default the Hotelling T-square chart is computed; select Comprehensive results to compute the chart, covariance matrix, and results spreadsheet with chart values; request All results to compute descriptive statistics, the inverse covariance matrix, etc.
Input data When multiple observations per sample is selected, it expects samples with more than one case; the T-square chart will then be based on the average within-sample variance/covariance matrix. When individual observations is selected, the T-square chart will be computed based on the overall variance/covariance matrix.
Phase of control
Number of obs. per sample Number of observations per sample; each consecutive set of k observations will be assigned to the same sample; if a categorical predictor is specified, then the (first) categorical predictor variable will be used to assign observations to samples.
Alpha probability Specify the p-value (alpha probability) to compute the UCL or upper control limit in the Hotelling T-Square chart.
Moving average line The user can include a moving average line in the T-square chart.
N-point moving average; N Specify the number of samples (n) that are to be averaged for each point in the standard moving average line.
Exponentially weighted moving average; L Specify the exponent for the exponentially weighted moving average. This moving average line smoothes the pattern of variables (rates, proportions) across samples, and allows the engineer to detect trends more easily.
Save within group std. dv's & correlations To save the current matrix of correlations between the variables selected for the plot.