Step1(2). Seed column usage
Oracle must observe a representative workload, in order to determine the appropriate column groups. Using the new procedure DBMS_STATS.SEED_COL_USAGE, you tell Oracle how long it should observe the workload.
Step 2: (3) You don't need to execute all of the queries in your work during this window.
You can simply run explain plan for some of your longer running queries to ensure column group information is recorded for these queries.
Step 3. (1) Create the column groups
At this point you can get Oracle to automatically create the column groups for each of the tables based on the usage information captured during the monitoring window. You simply have to call the DBMS_STATS.CREATE_EXTENDED_STATS function for each table.This function requires just two arguments, the schema name and the table name. From then on, statistics will be maintained for each column group whenever statistics are gathered on the table.
Note:
*DBMS_STATS.REPORT_COL_USAGE reports column usage information and records all the SQL operations the database has processed for a given object.
*The Oracle SQL optimizer has always been ignorant of the implied relationships between data columns within the same table. While the optimizer has traditionally analyzed the distribution of values within a column, he does not collect value-based relationships between columns.
*Creating extended statisticsHere are the steps to create extended statistics for related table columns withdbms_stats.created_extended_stats:
1 - The first step is to create column histograms for the related columns.
2 - Next, we run dbms_stats.create_extended_stats to relate the columns together.
Unlike a traditional procedure that is invoked via an execute ("exec") statement, Oracle extended statistics are created via a select statement.

Explanation/Reference:
Explanation:
E: You can perform all ALTER PLUGGABLE DATABASE tasks by connecting to a PDB and running the corresponding ALTER DATABASE statement. This functionality is provided to maintain backward compatibility for applications that have been migrated to a CDB environment.
A, G: When you issue an ALTER PLUGGABLE DATABASE OPEN statement, READ WRITE is the default unless a PDB being opened belongs to a CDB that is used as a physical standby database, in which case READ ONLY is the default.
You can specify which PDBs to modify in the following ways:
List one or more PDBs.

Specify ALL to modify all of the PDBs.

Specify ALL EXCEPT to modify all of the PDBs, except for the PDBs listed.

http://docs.oracle.com/database/121/NETAG/listenercfg.htm

* (C,E,not a) file_name_convert
Use this clause to determine how the database generates the names of files (such as data files and wallet files) for the PDB.
For filename_pattern, specify a string found in names of files associated with the seed (when creating a PDB by using the seed), associated with the source PDB (when cloning a PDB), or listed in the XML file (when plugging a PDB into a CDB).
For replacement_filename_pattern, specify a replacement string.
Oracle Database will replace filename_pattern with replacement_filename_pattern when generating the names of files associated with the new PDB.
File name patterns cannot match files or directories managed by Oracle Managed Files.
You can specify FILE_NAME_CONVERT = NONE, which is the same as omitting this clause. If you omit this clause, then the database first attempts to use Oracle Managed Files to generate file names. If you are not using Oracle Managed Files, then the database uses the PDB_FILE_NAME_CONVERT initialization parameter to generate file names. If this parameter is not set, then an error occurs.
Note:
* Oracle Database 12c Release 1 (12.1) introduces the multitenant architecture. This database architecture has a multitenant container database (CDB) that includes a root container, CDB$ROOT, a seed database, PDB$SEED, and multiple pluggable databases (PDBs).