Exercises

[maria65mj]

No description provided.

[ivanflee]

Not sure if I am correct, but based on the proposed output, the region_name needs to be aggregated using the string_agg() function and not out on its own. And then we can group by the growth (what you have as 'comparison_growth'.

Based on your code it would look something like this:

SELECT  
      (CASE WHEN (y2016::float/y1985)>=3.652 THEN 'High Growth' ELSE 'Low Growth' END) as comparison_growth,
      STRING_AGG(region_name,', ') 
FROM (SELECT region_name, 
      sum(CASE WHEN year=1985 THEN gdp_per_capita ELSE NULL END) AS y1985,
      sum(CASE WHEN year=2016 THEN gdp_per_capita ELSE NULL END) AS y2016
      FROM public.gdp
      GROUP BY region_name) as original
WHERE y2016 IS NOT NULL AND y1985 IS NOT NULL
GROUP BY comparison_growth;