ID 	Name	Salary	Data
-------------------------
1	Test1 	1000	201401
2	Test2	2000	201501
3	Test3	3000	201301
4	Test4	4000	
10	Test10	500	200101

Seq_no 	Name	ID	Salary		Data
---------------------------------------
100		Test11	1	2000		201612
101		Test11	1	2000		201401
102		Test22	2	1050		201501
103		Test22	2	1050		200108
104		Test44	4	3500		200002
105		Test55	5  	5500		201205
106		Test66	6	1200		201201
107		Test10	10	700

ID 	Name	Salary	Data
-------------------------
1	Test11 	2000	201612			- Data picked from lower deq_num record
2	Test22	1050	201501			- Data picked from lower deq_num record
3	Test33	3000	201301
4	Test44	3500	200002			- Data updated with new record	
5	Test55	5500	201205	
6	Test66	1200	201201	
10	Test10	700	200101			- T2 has null but data is retained as in T1

merge into t1 o
using (select rowid as rid, 
			id as i_id,
			name as i_name,
			salary as i_salary,
			data as i_date
		from t2 
        where rowid in (select max(rowid) from t2 group by indiv_id))
on (o.id = i_id)
WHEN MATCHED THEN
UPDATE
SET o.name = i_name,
	o.salary = i_salary,
	o.data = i_date
When not matched then
insert (id, name, salary, data)
values (i_id, i_name, i_salary, i_data); 

MERGE INTO T1 i
USING ( 
    select distinct o1.id as o_id,
			o1.data as o_data
    from T2 o1 join T2 o2
    on (o1.id = o2.id 
        and nvl(o1.data,'0') > nvl(o2.data,'0'))
    ) o
ON (i.indiv_id = o_id)
WHEN MATCHED THEN 
UPDATE SET 
    i.data =  nvl(o_data,i.data);

create table T1
(id        number,
 name    varchar2(100),
 salary    number,
 data   varchar2(6) );
 
 ALTER TABLE T1 ADD PRIMARY KEY (id);
 
insert into T1 values(1 , 'Test1' ,    1000, '201401');
insert into T1 values(2    , 'Test2' ,    2000, '201501');
insert into T1 values(3    , 'Test3' ,    3000, '201301');
insert into T1 values(4    , 'Test4' ,    4000, NULL);
insert into T1 values(10, 'Test10',    500    , '200101');

create table T2
(seq_no    number,
 id        number,
 name    varchar2(100),
 salary    number,
 data   varchar2(6));
 
insert into T2 values (100, 1 ,'Test11', 2000, '201612');
insert into T2 values (101, 1 ,'Test11', 2000, '201401');
insert into T2 values (102, 2 ,'Test22', 1050, '201501');
insert into T2 values (103, 2 ,'Test22', 1050, '200108');
insert into T2 values (104, 4 ,'Test44', 3500, '200002');
insert into T2 values (105, 5 ,'Test55', 5500, '201205');
insert into T2 values (106, 6 ,'Test66', 1200, '201201');
insert into T2 values (107, 10,'Test10', 700 , NULL);

ID 	Name	Salary	Data
-------------------------
1	Test11 	2000	201612			- Data picked from lower seq_no record but with highest "data" value for the ID 1
2	Test22	1050	201501			- Data picked from lower seq_no record but with highest "data" value for the ID 1
3	Test33	3000	201301
4	Test44	3500	200002			- Data updated with new record	
5	Test55	5500	201205	
6	Test66	1200	201201	
10	Test10	700	200101			- T2 has null but data is retained as in T1

Select id
     , seq_no
     , salary
     , name
     , data
From   ( select id
              , seq_no
              , salary
              , name
              , data
              , dense_rank() over ( partition by id order by data desc) dr
         from   t2
       )

SQL> DROP TABLE T1 PURGE;

Table dropped.

SQL>
SQL> CREATE TABLE T1
  2    (id NUMBER, name VARCHAR2(100), salary NUMBER, dt DATE
  3    );

Table created.

SQL> ALTER TABLE T1 ADD PRIMARY KEY
  2  (
  3    id
  4  )
  5  ;

Table altered.

SQL>
SQL> INSERT INTO T1 VALUES
  2    (1 , 'Test1' , 1000, TO_DATE('201401','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T1 VALUES
  2    (2 , 'Test2' , 2000, TO_DATE('201501','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T1 VALUES
  2    (3 , 'Test3' , 3000, TO_DATE('201301','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T1 VALUES
  2    (4 , 'Test4' , 4000, NULL
  3    );

1 row created.

SQL> INSERT INTO T1 VALUES
  2    (10, 'Test10', 500 , TO_DATE('200101','YYYYMM')
  3    );

1 row created.

SQL>
SQL> DROP TABLE T2 PURGE;

Table dropped.

SQL>
SQL> CREATE TABLE T2
  2    (
  3      seq_no NUMBER,
  4      id     NUMBER,
  5      name   VARCHAR2(100),
  6      salary NUMBER,
  7      dt     DATE
  8    );

Table created.

SQL>
SQL> INSERT INTO T2 VALUES
  2    (100, 1 ,'Test11', 2000, TO_DATE('201612','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T2 VALUES
  2    (101, 1 ,'Test11', 2000, TO_DATE('201401','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T2 VALUES
  2    (102, 2 ,'Test22', 1050, TO_DATE('201501','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T2 VALUES
  2    (103, 2 ,'Test22', 1050, TO_DATE('200108','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T2 VALUES
  2    (104, 4 ,'Test44', 3500, TO_DATE('200002','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T2 VALUES
  2    (105, 5 ,'Test55', 5500, TO_DATE('201205','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T2 VALUES
  2    (106, 6 ,'Test66', 1200, TO_DATE('201201','YYYYMM')
  3    );

1 row created.

SQL> INSERT INTO T2 VALUES
  2    (107, 10,'Test10', 700 , NULL
  3    );

1 row created.

SQL>

SQL> COLUMN NAME FORMAT A10;
SQL>
SQL> SELECT val id,
  2    name,
  3    salary,
  4    TO_CHAR(dt, 'YYYYMM') max_dt
  5  FROM
  6    (SELECT ID val ,
  7      NAME,
  8      SALARY,
  9      DT
 10    FROM T2
 11    JOIN
 12      (SELECT ID val, MAX(DT) MAX_DT FROM T2 GROUP BY ID
 13      ) A
 14    ON (t2.dt = a.max_dt)
 15    )
 16  ORDER BY id
 17  /

        ID NAME           SALARY MAX_DT
---------- ---------- ---------- ------
         1 Test11           2000 201612
         2 Test22           1050 201501
         4 Test44           3500 200002
         5 Test55           5500 201205
         6 Test66           1200 201201

SQL>

SELECT name,
	id     ,
	salary ,
	data   
FROM (SELECT o.*,
	ROW_NUMBER() OVER (PARTITION BY id ORDER BY seq_num DESC NULLS LAST) RN
FROM T2 o)
WHERE RN = 1;