jeudi 4 janvier 2018

Sort column within a table into 2 random buckets, 1000 times

I want to randomly bucket a list into 2 groups 1000 times

There is a list of 217 cities, that I want to bucket into a test group and a control group a lot of times, and then plot the ratio in a count per day for the two groups to see what sort of variance I can expect.

Table structure(date, grouping, count)

1   2017-01-01  Abilene-Sweetwater-TX   42
2   2017-01-01  Albany-GA   38
3   2017-01-01  Albany-Schenectady-Troy-NY  131
4   2017-01-01  Albuquerque-SantaFe-NM  241
5   2017-01-01  Alexandria-LA   22

.......

Ultimately I want to have a plot of 1000 lines, that is the ratio of counts for the test and control groups for each random bucketing scenario.




Aucun commentaire:

Enregistrer un commentaire