Accessing the results of a partial adjustment

 

Consider the following scenario: Suppose that base has been adjusted initially by base and then partially by base . Suppose that we might observe and (otherwise, data-related results are not available).

Size of a partial adjustment

 

Usage

• psize  operand.

If data is available, the size of a partial adjustment (the variance of the bearing for the partial adjustment) is returned, that is . See [31, section 5.4,].  

Discrepancy for a partial adjustment

 

Usage

• dvpsize  operand.

If data is available, the discrepancy for a partial adjustment (the largest squared change in expectation, relative to the amount of prior variation removed by the partial adjustment) is returned, that is .  

Partial resolution matrix rank

 

Usage

• prmrank  operand.

The rank of the partial resolution matrix, being the number of partial canonical directions corresponding to non-zero partial canonical resolutions. See [31, section 5.3,].  

Partial resolution matrix trace

 

Usage

• prmtr  operand.

The trace of the partial resolution matrix, equal to the sum of the partial canonical resolutions. See [31, section 5.3,].  

Partial canonical resolutions

 

Usage

• pcr  operator; one argument i in parenthesis. i is any valid equation, rounded to the nearest integer.

This returns the ith largest partial canonical resolution. It is an error if the index i is smaller than one or greater than the rank of the partial resolution matrix (obtainable as the operand prmrank ). For example, suppose that is the 4th partial canonical direction for the adjustment of by given . Then . See [31, section 5.3,]  

Path correlation

 

Usage

• pc  operand.

This returns the current path correlation, , or zero if it is undefined. See [31, section 5.5,].