Data Science and Big data, SQL

  

1. How would you use GROUPING SETS to produce the same results as the following GROUP BY CUBE?
SELECT state, productID, SUM(volume) FROM sales GROUP BY CUBE (state, productID) ORDER BY state, productID
2. How would you show the sub totals for each week, for each state, and for each product? (No other totals or grand totals are required.) Suppose the table structure is
TABLE sales (productID VARCHAR, state CHAR(2), week DATE, volume INT)
3. Discuss the utility of grouping and group_id functions
Q2: Describe below how this code will work:

SELECT dt, region, revenue,

count(*) OVER (twdw) AS moving_count,

avg(revenue) OVER (twdw) AS moving_average

FROM moving_average_data mad

WINDOW twdw AS (PARTITION BY region

ORDER BY dt RANGE BETWEEN

‘7 days’::interval PRECEDING AND

‘0 days’::interval FOLLOWING

ORDER BY region, dt