...

Select variables using p-value

 

The method of statistical hypothesis testing can also be used to determine whether the independent variable has a significant impact on the dependent variable. SPL provides several functions for statistical testing p-value calculation. Function usage: p value (raqsoft.com).

In this case, variables in credit card data were selected in the form of T-test, and the screening criteria was to retain variables with a p-value less than 0.01.

A

B

C

1

=file("D://test//creditcard_b.csv").import@tc()

2

=A1.fname()

3

=A2.delete(A2.pos("Class"))

4

for A2

=ttest_p(A1.(${A4}),A1.(Class))

5

>B1=B1|[A4|B4]

6

=if(B4<0.01,A4)

7

>C1=C1|B6

A2 Get field names

A3 Deletes the target field name

A4-B7 Loop each field, calculate the p-value of each independent variable and target variable respectively and put the results into B1, and filter the variables with a P-value less than 0.01 into C1

.. ..


 
 •  160 View  •