7.00.02 - Scale Arguments - Aster Analytics

Teradata Aster® Analytics Foundation User GuideUpdate 2

Product
Aster Analytics
Release Number
7.00.02
Release Date
September 2017
Content Type
Programming Reference
User Guide
Publication ID
B700-1022-700K
Language
English (United States)
Method
Specifies one or more statistical methods to use to scale the data set. For method values and descriptions, see the table "Location and Scale for Statistical Methods."

If you specify multiple methods, the output table includes the column scalemethod (which contains the method name) and a row for each input-row/method combination.

Global
[Optional] Specifies whether all input columns are scaled to the same location and scale. For a description of location and scale and the values that are used with each Method, see the table "Location and Scale for Statistical Methods." Default: 'false' (each input column is scaled separately).
InputColumns
[Optional] Specifies the input table columns that contain the values to scale. The columns must contain numeric values between -1e308 and 1e308. Default: All columns of the statistic table (the output of the ScaleMap function in the Scale syntax) except stattype.
Accumulate
[Optional] Specifies the input table columns to copy to the output table.
Multiplier
[Optional] Specifies one or more multiplying factors to apply to the input variables—multiplier in the following formula:

X' = intercept + multiplier * (X - location)/scale

If you specify only one multiplier, it applies to all columns specified by the InputColumns argument. If you specify multiple multiplying factors, each multiplier applies to the corresponding input column. For example, the first multiplier applies to the first column specified by the InputColumns argument, the second multiplier applies to the second input column, and so on. Default: multiplier is 1.

Intercept
[Optional] Specifies one or more addition factors incrementing the scaled results—intercept in the following formula:

X' = intercept + multiplier * (X - location)/scale

If you specify only one intercept, it applies to all columns specified by the InputColumns argument. If you specify multiple addition factors, each intercept applies to the corresponding input column.

The syntax of intercept is:

[-]{number | min | mean | max }

where min, mean, and max are the scaled global minimum, maximum, mean values of the corresponding columns.

The formula for computing the scaled global minimum is:

scaledmin = (minX - location)/scale

The formulas for computing the scaled global mean and maximum are analogous to the preceding formula.

For example, if intercept is '- min' and multiplier is 1, the formula for computing the scaled result X' from the original value X is:

X' = -scaledmin + 1 * (X - location/scale)

Default: intercept is 0.

Two important parameters in scaling are location and scale. The values that the function uses for location and scale depend on the Method argument value, as the following table shows. In the table, X represents an input column, and Xmean, minX, and maxX are the mean, minimum, and maximum values of the column X.

Location and Scale for Statistical Methods
Method Location Scale
mean X mean 1
sum 0 Σ X
ustd 0 Standard deviation, calculated according to the biased estimator of the variance.
std X mean Standard deviation, calculated according to the unbiased estimator of the variance.
range minX maxX - minX
midrange (maxX+minX)/2 (maxX - minX)/2
maxabs 0 Maximum of the absolute value of X.