Hi Sudarshan,
Although I did not try this myself, can you please try using CE_AGGREGATION instead of CE_PROJECTION for SQL_P.
As the aggregation depends on the grouping, if you have a lower granularity data for any other attribute along with COLLAB_ID, you may not get the correct result. In that case, you may have to create another SQL_ view just for COLLAB_ID and max(datetimestamp) and then use CE_JOIN to connect it with the rest of the data.
Regards,
Ravi