You design the data migration process. The legacy data records have ownership problems. For some records, the ownership column is empty. For others — store owners that are no longer with the company. 
Please select three methods you can use to correct this problem.

Answer options:

A.Use stub users
B.Use service identity
C.Use an empty column
D.Use predefined user
E.Skip the column and provision a new column after migration

Correct Answers: A, B and
D. 
When a Solution Architect designs the data migration, he or she needs to consider several things, like compliance, lack of documentation for the legacy systems and data, data duplication, required columns are empty, no unique record identifiers, no established relationships between tables, missing ownership of the records, etc. 
The Dataverse requires record ownership. If the legacy data cannot provide this information, you need to work with stakeholders to solve this problem. There are three methods: Use a predefined user — the stakeholders can provide a user, fully registered in the Dataverse, that will be used in case of missing values or for all the records.
Use a stub user — this is a user who does not have a Dataverse login or data access but can be used only to import the legacy data into the Dataverse and preserve the original record owners. You can use this user type only for Excel-based import.
Use a service identity — the identity should be provisioned in the Dataverse and have all the security rights to write data to all Dataverse tables and perform the API calls. This identity will own all the records without a specific owner.
All other options are incorrect. 
For more information about the data migration planning, please visit the below URLs: 

https://docs.microsoft.com/en-us/learn/modules/project-test-process/3-migrate-data
https://docs.microsoft.com/en-us/powerapps/developer/data-platform/import-data