Statistical Functions - trillium_discovery - Latest

An outlying value is one that is outside the standard range of data. Use the results of the STDDEV function to determine the deviation level to use.

The function returns true (1) for all numeric data whose deviation is equal to or greater than the deviation level specified with the function. The function returns false (0) for all numeric values with deviation less than the deviation level specified with the function.

Note the following guidelines:

• Function is not supported for Quality processes.
• Function is not supported for dynamic entities (data sources). Applicable only to real, fully-loaded entities (data sources).
• You must run Std. Deviation analysis on the attribute before you use this function.

Syntax

ISOUTLIER(value,deviation_level)

Where

• value is either a number or an attribute.
• deviation_level is the absolute deviation level of 0, 1, 2, and so on through to the maximum value found in the data. The higher the number, the more outside the standard range of data the outlier is considered.

Example

For ISOUTLIER(att7,1), if att7 contains the values 3, 25, 27, 28, 29, 32, 33, and 85, returns true (1) for 3 and 85 and false (0) for the other rows.

For the same set of data, ISOUTLIER(att7,0) returns true (1) for all rows.

Mean returns the numeric average of the values in an attribute, determined by the sum of values in the attribute divided by the number of values.

Note the following guidelines:

• Function is not supported for Quality processes.
• Function is not supported for dynamic entities (data sources). Applicable only to real, fully-loaded entities (data sources).

Syntax

MEAN(attribute)

Where attribute is the attribute that contains the data for which you want to calculate the average.

Example

For MEAN(att1), if att1 contains the values 3, 25, 27, 28, 29, 32, 33, 85 the mean will be 32.75.

Median is the numeric value separating the higher half of a data set from the lower half. This function returns the median of the specified attribute.

Note the following guidelines:

• Function is not supported for Quality processes.
• Function is not supported for dynamic entities (data sources). Applicable only to real, fully-loaded entities (data sources).
• You must run percentile analysis on the attribute before you use this function.

Syntax

MEDIAN(attribute)

Where attribute is the set of data for which you want to find the median value.

Example

For MEDIAN (att20), if att20 contains the values 1 2 3 4 3 2 1 2 3 5 6 1, 3 will be returned.

Returns the rank of a specified numeric value in a data set. The percentrank represents the location of the percentile in a value. For example, halfway in a value the percentrank would be 50%. Percentrank is calculated to two decimal points.

Note the following guidelines:

• Function is not supported for Quality processes.
• Function is not supported for dynamic entities (data sources). Applicable only to real, fully-loaded entities (data sources).
• You must run percentile analysis on the attribute before you use this function.

Syntax

PERCENTRANK(attribute,value)

Where

• attribute is the attribute that contains the data in which you want to rank the value.
• value is the number whose rank you want to find.

Example

For PERCENTRANK(att5,3), if att5 contains the values 1 2 3 4 3 2 1 2 3 5 6 1 the percentrank will be 54.55%.

A quartile is used to rank numeric data into quarters. The function returns the requested quartile of a value in an attribute as a percentage of the attribute.

Note the following guidelines:

• Function is not supported for Quality processes.
• Function is not supported for dynamic entities (data sources). Applicable only to real, fully-loaded entities (data sources).
• You must run percentile analysis on the attribute before you use this function.

Syntax

QUARTILE(attribute,value)

Where

• attribute is the name of the attribute that contains value.
• value is the value used to determine the quartile percentage. Supported values include:
  • 0. Returns the minimum value in the attribute.
  • 1. Returns the first number with percent rank equal to or greater than 25.
  • 2. Returns the first number with percent rank equal to or greater than 50.
  • 3. Returns the first number with percent rank equal to or greater than 75.
  • 4. Returns the maximum value in the attribute.

Examples

• For QUARTILE(att10,0), if att10 contains the values 1 2 3 4 3 2 1 2 3 5 6 1, 1 is returned.
• For the same set of data, QUARTILE(att10,4) would return 6.

Returns the standard deviation of an attribute, calculated to three decimal points. Standard deviation is used to measure how widely values are removed from the average value in an attribute. Use the results of this function to run the ISOUTLIER function.

Note the following guidelines:

• Function is not supported for Quality processes.
• Function is not supported for dynamic entities (data sources). Applicable only to real, fully-loaded entities (data sources).
• You must run Std. deviation analysis on the attribute before you use this function.

Syntax

STDDEV(attribute)

Where attribute is the attribute for which you want the standard deviation.

Example

For STDDEV(att2), if att2 contains the values 1 2 3 4 3 2 1 2 3 5 6 1, the standard deviation will be 1.534.

Returns the total of a specified attribute.

Note the following guidelines:

• Function is not supported for Quality processes.
• Function is not supported for dynamic entities (data sources). Applicable only to real, fully-loaded entities (data sources).

Syntax

TOTAL(attribute)

Where attribute is the attribute whose numeric values you want to total.

Example

For TOTAL(att25), if att25 contains the values contains the values 1 2 3 4 3 2 1 2 3 5 6 1, the total returned will be 33.