optimize query

Greetings!

I have a following query:
as you see, i have joined atfirst one "item" id table with a auto generated table of dates.
as a result i got table where every item_id has assigned about 365 last dates.
After that i join it to a massive table which contains every day statistics for items. in the end query returns item_id's and dates where data is missing and is not present in statistics table.

I wish to know, is there anything i can do to optimize the query?
or is there another way how to get id's of missing items at certain dates?

item table has about 12 000 records;(after filtering)
statistic table has about 50 000 000 records

first time query call takes 160 sec, after cashing 20 secs ;
am using multicolumn index acording to all 3 fields in where clause on statistics table;

query plan:
"Aggregate  (cost=375007.39..375007.40 rows=1 width=0)"
"  ->  Nested Loop Anti Join  (cost=0.01..375007.39 rows=1 width=0)"
"        ->  Nested Loop  (cost=0.00..3130.29 rows=30000 width=8)"
"              ->  Seq Scan on preces  (cost=0.00..2455.29 rows=30 width=4)"
"                    Filter: (sort_a AND (NOT stop_prece) AND (id = id))"
"              ->  Function Scan on generate_series a  (cost=0.00..12.50 rows=1000 width=4)"
"        ->  Index Scan using realizacija_pkey on realizacija  (cost=0.01..12.90 rows=1 width=8)"
"              Index Cond: ((realizacija.veikala_id = 1) AND (preces.id = realizacija.prece_id) AND (realizacija.datums = (('now'::text)::date - a.a)))"

database type: Postgres

select count(*) 
from(
     select * 
     from(
               select current_date - a as dat from generate_series(2,370) as a) Q1
          join(
               select id, artikuls  from preces where sort_a=true and stop_prece=false) Q2 
          on q2.id=q2.id
          
     left join(
              select prece_id,datums
              from realizacija 
              where veikala_id=1) Q3 
     on q3.datums=q1.dat and q2.id=q3.prece_id) as Q1 
where datums is null

Select all Open in new window