Jrb5Wje7

1create table if not exists my_timeseries as
2       select
3           coalesce(sector, district, area) as postcode,
4           import_date,
5           'A'::character(1) as property_type,
6           -1::smallint as bedrooms,
7
8           CAST (sum(1) filter(where is_begin_ts) AS REAL) as s_listings,
9           CAST (percentile_cont(array[0.25,0.5,0.75,0.9]) within group (order by asking_price) filter(where asking_price > 0) AS REAL[]) as s_price_asked
10
11       from my_source_table
12       where import_date <= '2019-05-01'::date
13       grouping sets ((sector, import_date), (district, import_date), (area, import_date));