Reputation: 4099
Consider the following table:
sp100_id _date open close bullishness
-----------------------------------------------
1 2011-03-14 100 110 2.23
1 2011-03-15 115 100 1.00
1 2011-03-16 110 110 0.85
2 2011-03-14 90 85 0.99
2 2011-03-15 95 90 0.30
2 2011-03-16 92 100 4.66
3 2011-03-14 200 220 1.50
3 2011-03-15 250 210 1.75
3 2011-03-16 200 150 0.80
I want to calculate the % return
and bullishness
for each stock between 2011-03-14
and 2011-03-16
, rank them by % return
DESC, and the return sp100_id
, % return
and average bullishness
. I thought the following query would do the trick:
SELECT
sp100_id,
AVG(bullishness) as bullishness,
((`close`-`open`) / `open`) as return_pct
FROM
stocks
WHERE _date = BETWEEN '2011-03-14' AND '2011-03-16'
ORDER BY return_pct DESC
However, it always just seems to return just one row. The expected output was however:
sp100_id return_pct average bullishness
-----------------------------------------------------------
3 (150-200)/200 = -0.250 (1.50+1.75+0.80)/3 = 4.05
2 (100-90)/90 = 0.110 (0.99+0.30+4.66)/3 = 1.98
1 (110-100)/100 = 0.100 (2.23+1.00+0.85)/3 = 4.08
What am I doing wrong?
Upvotes: 2
Views: 126
Reputation: 11393
You must use GROUP BY
in your query:
SELECT
sp100_id,
AVG(bullishness) as bullishness,
(((SELECT `close` FROM stocks s2
WHERE s2.sp100_id=s1.sp100_id ORDER BY _date DESC LIMIT 1)
-(SELECT `open` FROM stocks s3
WHERE s3.sp100_id=s1.sp100_id ORDER BY _date ASC LIMIT 1)) /
(SELECT `open` FROM stocks s3
WHERE s3.sp100_id=s1.sp100_id ORDER BY _date ASC LIMIT 1)) as return_pct
FROM
stocks s1
WHERE _date BETWEEN '2011-03-14' AND '2011-03-16'
GROUP BY sp100_id
ORDER BY return_pct DESC
Without GROUP BY
, it returns only one result row because AVG
groups all records together.
I fixed the computation of return_pct using subqueries with ORDER BY
and LIMIT 1
.
Documentation: GROUP BY (Aggregate) Functions
Upvotes: 3