Hands-On SAS for Data Analysis

Program data vector

We explored the concept of the Program Data Vector (PDV) in previous chapters. In this section, we will look at what the PDV of the previously merged dataset would look like. This will provide us with a further understanding of how the Merge process works. We will look at the roles of _N_ and _ERROR_. The focus is on understanding how observations from the three datasets will consolidate into one dataset. The following are the steps that PDVs follow:

1. Initially, the PDV will be created by incorporating the variables from datasets A, B, and C and by setting all the records to missing:

City	Index	Housing	Food	Utility	Education	Index_Date
	.	.	.	.	.	.

2. The program will look for the first ...