Transformations to Y, Columns Variables

The following options specify the form of the Y, Columns variables to be used in the cluster analysis:

Addresses the issue of different measurement scales for continuous and ordinal columns. Except when the Data is stacked option is selected, the values in each column are standardized by subtracting the column mean and dividing by the column standard deviation. Deselect the Standardize Data check box if you do not want the cluster distances computed on standardized values.

Standardize Robustly

Reduces the influence of outliers on estimates of the mean and standard deviation for continuous and ordinal columns. This option uses Huber M-estimates of the mean and standard deviation (Huber 1964; Huber 1973; Huber and Ronchetti 2009). For columns with outliers, this option gives the standardized values greater representation in determining multivariate distances.

Note: If both Standardize Data and Standardize Robustly are selected, each column is standardized by subtracting its robust column mean and dividing by its robust standard deviation. This option is useful when columns represent different measurement scales or when observations tend to be outliers in only specific dimensions.

Note: If Standardize Data is unchecked and Standardize Robustly is selected, the robust mean and robust standard deviation for the values in all columns combined are used to standardize each column. This option can be useful when columns all represent the same measurement scale and when observations tend to be outliers in all dimensions.

Missing value imputation

Imputes missing values. If the number of variables is either 50 or less, or less than half the number of rows, multivariate normal imputation is used. Otherwise, multivariate SVD imputation is used.

Multivariate normal imputation calculates pairwise covariances to construct a covariance matrix for the response columns. Then each missing value is imputed by a method that is equivalent to regression prediction using all the predictors with no missing values for the given observation. If the constructed covariance matrix is not positive definite, missing values are imputed using their column means.

Multivariate SVD imputation avoids constructing a covariance matrix by using the singular value decomposition. For more details, see Explore Missing Values Utility in the Predictive and Specialized Modeling book.

Caution: Missing value imputation assumes that there are no clusters, that the data come from a single multivariate normal distribution, and that the values are missing completely at random. Because these assumptions are usually not reasonable in practice, use this feature with caution. However, the feature can produce more informative results than discarding most of your data.

Add Spatial Measures

(Available only if Data is stacked is selected as the data structure.) Select the Add Spatial Measures option when your data are stacked and contain two attribute columns that correspond to spatial coordinates (horizontal and vertical coordinates, for example). This option opens a window in which you can select and weight spatial components to aid in clustering defect patterns. This is a specialty method and is applicable in only very specific settings. See Spatial Measures and Example of Wafer Defect Classification Using Spatial Measures.