I have the following table:

"CREATE TABLE IF NOT EXISTS daily( id INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL UNIQUE," "value REAL," "date_id INTEGER NOT NULL," "symbol_id INTEGER NOT NULL," "FOREIGN KEY (symbol_id) REFERENCES sym_master (symbol_id)" ")"

I insert values :

INSERT INTO daily (value, date_id, symbol_id) VALUES (?,?,?)", [5.99, 1, 1])

The table looks like this:

value, date_id, symbol_id
[5.99, 1, 1]
[5.99, 2, 1]
[5.99, 3, 1]
.
.
.
[5.99, 1, 2]
[5.99, 2, 2]
[5.99, 17, 2]
.
.
.

The date_id will have a max value for every group of the symbol_id

-> lets say the max Value for every entry with symbol_id == 1 is 2314, and the max Value for all entries which have a symbol_id of 2 is 1326. There are many symbol_id 1 to n

i want to query the table, and for every unique symbol_id in it, getting the MAX(date_id) value, so i would return something like:

symbol_id Max(date_id) 1 2314 2 1326 3 . . . n` ...

SELECT DISTINCT symbol_id, MAX(date_id) FROM daily

gives me 'only' the MAX Value of the whole column, i would like to have it by group, can someone point me to the right approach?