A.mvp()

Description:

Create indicator variables for the MVP analysis and automatically perform the subsequent handling according to a sequence of indicator variables for missing values.

Syntax:

A.mvp(T)

During modeling, create indicator variables for MVP analysis and automatically perform the subsequent handling according to sequence A of multiple missing value indicator variables, and return a binary sequence consisting of two bits – the first bit represents a table sequence made up of MVP columns and the second one is sequence Rec of creation and handling records

A.mvp@r(Rec)

During scoring, create a table sequence consisting of MVP columns according to sequence A of multiple missing value indicator variables and sequence Rec of creation and handling records

Note:

The external library function (See External Library Guide) creates indicator variables for the MVP analysis and automatically perform the subsequent handling according to a sequence of indicator variables for missing values.

Parameter:

A

A sequence

T

A sequence of target variable values

Rec

A sequence of creation and handling records

Option:

@bnie

Each option specifies a target type, and options are mutual-exclusive; automatically judge the type when no option is used; the order of priorities of the options is binary, numeric, integer and enumerated

Return value:

A sequence

Example:

 

 

A

 

1

=T("D://house_prices_train.csv")

Import modeling data.

2

=T("D://house_prices_test.csv")

Import scoring data.

3

=A1.mi("LotFrontage").field(1)

Return a sequence of indicator variables for missing values in “LotFrontage” .

4

=A1.mi("Alley").field(1)

Return a sequence of indicator variables for missing values inf “Alley”.

5

=A1.(SalePrice)

A sequence of target variable values.

6

=[A3,A4].mvp(A5)

A6(1): A table sequence containing MVP columns;

A6(2):A sequence of creation and handling records Rec.

7

=A2.mi("LotFrontage").field(1)

 

8

=A2.mi("Alley").field(1)

 

9

=[A7,A8].mvp@r(A6(2))

Return a table sequence made up of indicator variables of MVP columns for the scoring.