Home/ Questions/Q 8322969
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-08T23:33:54+00:00

(This is quite a long post, but the problem is I think easy to

  • 0

(This is quite a long post, but the problem is I think easy to solve and I have a SQLFiddle ready) Please consider the following table:

----------------------------------------------------------------------
tweet_id sp100_id nyse_date   user_id class_id retweets quality follow
----------------------------------------------------------------------
1        1        2011-03-12  1       1        0        2.50    5.00
2        1        2011-03-13  1       2        2        2.50    5.00
3        1        2011-03-13  1       2        1        2.50    5.00
4        1        2011-03-13  2       2        0        0.75    1.00
5        1        2011-03-13  2       3        3        0.75    1.00
6        2        2011-03-12  2       2        3        0.75    1.00
7        2        2011-03-12  2       2        0        0.75    1.00
8        2        2011-03-12  1       3        5        2.50    5.00
9        2        2011-03-13  2       2        0        0.75    1.00
----------------------------------------------------------------------

The desired output from this table is a list per sp100_id per _date the amount of positive (class=2) and negative (class=3) tweets weighted per retweets, quality and follow:

--------------------------------------------------------------------------------
sp100_id  nyse_date  pos-rt pos-quality pos-follow neg-rt neg-quality neg-follow
--------------------------------------------------------------------------------
1         2011-03-11 0      0           0          0      0           0
1         2011-03-12 0      0           0          0      0           0
1         2011-03-13 3 (1)  5.75 (2)    11.00 (3)  3 (4)  0.75        1.00
2         2011-03-11 0      0           0          0      0           0
2         2011-03-12 3      1.50        10.00      5.00   2.50        2.50
2         2011-03-13 0      0.75        1.00       0      0           0
--------------------------------------------------------------------------------

On 2011-03-13, 3 positive tweets for sp100_id 1:

(1) 1 tweet retweeted 2 times, 1 tweets retweeted 1 time and 
    1 tweet retweeted 0 times = 1 x 2 + 1 x 1 + 1 x 0 = 3
(2) 2 tweets with quality 2.50 and 1 tweet with quality 0.75 =
    2 x 2.50 + 1 x 0.75 = 5.75
(3) 2 tweets with follow 5 and 1 tweet with follow 1 =
    2 x 5.00 + 1 x 1.00 = 11.00

On 2011-03-13, 1 negative tweets for sp100_id 1:

(4) 1 tweet retweeted 3 times = 1 x 3 = 3

etc...

I have a demo on SQLFiddle with the necessary other tables (I need to link it to a daterange table because I also want to include recordsets with all zero’s). I also have an output for my query, but I don’t understand why it is different from the desired output:

--------------------------------------------------------------------------------
sp100_id  nyse_date  pos-rt pos-quality pos-follow neg-rt neg-quality neg-follow
--------------------------------------------------------------------------------
1         2011-03-11 0      0           0          0      0           0
1         2011-03-12 3      2           2          5      3           5
1         2011-03-13 3      8           12         3      1           1
2         2011-03-11 0      0           0          0      0           0
2         2011-03-12 3      2           2          5      3           5
2         2011-03-13 3      8           12         3      1           1
--------------------------------------------------------------------------------

I don’t see where the problem lies. Do you? Your help would be greatly appreciated 🙂

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    The reason why it wasn’t returning expected values is because you need to also include sp100.sp100_id = tweets.sp100_id in the LEFT JOIN condition along with the date.

    By only joining on the date, it will join on any date value in the table, regardless of sp100_id. This is why your resulting sums were being thrown off because for each sp100_id, it was including the values of all other sp100_ids in the SUM()s.

    I also cleaned up your query a little bit (just in terms of aesthetics):

    SELECT     a.sp100_id,
           b._date AS nyse_date,
           SUM(IF(c.class=2, c.retweets, 0)) AS 'pos-rt',
           SUM(IF(c.class=2, c.quality,  0)) AS 'pos-quality',
           SUM(IF(c.class=2, c.follow,   0)) AS 'pos-follow',
           SUM(IF(c.class=3, c.retweets, 0)) AS 'neg-retweet',
           SUM(IF(c.class=3, c.quality,  0)) AS 'neg-quality',
           SUM(IF(c.class=3, c.follow,   0)) AS 'neg-follow'
FROM       sp100 a
CROSS JOIN daterange b
LEFT JOIN  tweets c ON a.sp100_id = c.sp100_id 
                   AND b._date = c .nyse_date
GROUP BY   a.sp100_id, 
           nyse_date

    SQLFiddle Demo

    • 0