Workspace-Node-Specifications-for-Parameterization

    Transformation of Variable

    This section outlines the proper formatting for the Transformation parameter in Spotfire Analyst, especially when translating from Statistica. The unique formatting style involves placing each operator on a new line and separating comments with semicolons.

    Statistica Formatting

    In Statistica, the transformation parameter is formatted as follows:

    Spotfire Formatting

    To override the Transformation parameter in Spotfire, adhere to the following format:

    A=V1" "B=V2;This is comment" "C=Lookup(80,\"D:\Source\Statistica\Trunk\Debug-x64\Examples\Datasets\2level.sta\",1,0)

    Formatting Rules:

    1. Operators on New Lines:Each operator should be on a new line.

    2. Comments Separated by Semicolons: Comments should be separated by semicolons.

    3. Quoting Values:Quote each line using double quotes.

    4. Handling Inside Double Quotes:If inside a string, add \ at the start and end of the inside double quotes.

    PMML Node

    There are four parameters available on Spotfire

    Save Option:

    • +Copy

    • +Link

    • File Path

    • Enterprise Path

    To override parameter:

    • Save Option: Put the Copy or Link

    • File Path: Put the File path to local PMML file

    • Enterprise Path: If you want to use PMML store in Enterprise, Puth the Path here

    Sort Node

    Sorting involves four parameters:

    • Variables:

    This parameter represents the variable selected for sorting, consistent with other functioning nodes.

    • Sort By Array:

    This is a list of values that you want to sort by, whether numeric or textual.

    • Direction Array:

    This parameter is a list of directional values for sorting, indicating either ascending or descending order. It supports both "Ascending" and "Descending" values, with shorthand versions "Asc" and "Des."

    • Formatting:

    This parameter is a boolean value, accepting either true or false, indicating whether formatting should be applied during the sorting process.

    There is an example:

    • Variables: "TIME DEGREES YIELD"

    • Sort by Array: Numeric Text Numeric

    • Direction Array: Ascending

    • Formatting: False

    Process Missing Data

    • Input Variables: Selected Variables, Same with Variables of Sort node

    • Use case weights: bool variable(True/False)

    • Non-Numeric Data As MD: bool variable

    • White space as MD: bool variable

    • Recode Action: List of Recode Action, It could be: NoRecodeMD or RecodeMDToValue or RecodeMDToMean or RecodeMDToMedian RecodeMDToMode or FlagMD or RecodeMDToValueAndFlag

    Note:

    Note for Recode Action the FlagMD is currently available but it still does not work with the Spotfire Data table, So you can run it, but the value will not Flag in Spotfire. Maybe It will be supported in the future

    • Additional MD values: This is list of string, then It it will be separate by double quote and escape, like: “A” “B” “C”

    • Recode Value: Input same with Additional MD values

    • Flag if MD%: List of integer, separate by escape : ex: 1 0 0

    Record Outliers

  • Input Variables: Selected Variables, Same with Variables of Sort node

  • K interactions: Integer number

  • Repeat until all outliers have been recorded: bool

  • Measurement: list of measurement type( Unspecified, Auto, Continuous, Categorical, Ordinal) and the value separate by escape

  • Test: List of Test type:

    • Categorical

    • NormalTwoSided

    • NormalOneSidedUpper

    • NormalOneSidedLower

    • GrubbsTwoSided

    • GrubbsOneSidedUpper

    • GrubbsOneSidedLower

    • PercentileTwoSided

    • PercentileOneSidedUpper

    • TukeyTwoSided

    • PercentileOneSidedLower

    • TukeyOneSidedUpper

    • TukeyOneSidedLower

The value also separate by escape

  • Parameter: List of double value, each value also separated by escape (ex: 1 1.2 1.5)

  • Type: List of Recode type, There are the valid type:

    • NoRecode

    • RecodeToMD

    • RecodeToValue

    • RecodeToMean

    • RecodeToMode

    • RecodeToPercentile

    • RecodeToBoundary

The value also separate by Escape

  • Values: List of values in double, if there is any empty value it should be in double quote””

    For ex: "" "0.27"

  • Mark: List of integer

  • Use case weights: bool variable(True/False)

  • Case Selection Condition is apply for both Recode Outliers And Process Missing Data

  • Enable: This is bool value, If you want to use Case Selection Condition, this value should be true

  • Exclude Expression, Include Expression, Exclude List, Include List: there are the string field, It is follow other string field

  • Select All: This is bool value