Hi Devs I've started work on Z-Score reports but I'm having to use a javascript function to calculate standard deviation. I think we have a bug or user-error with the aggregationType=STDDEV implementation or I'm not using it correctly because I expect only 1 stDev value per dx based on the orgunit 'population' specified as criteria. This is how I would expect the API call to work:
api/analytics.json?*dimension*=dx:Ar2V5v1R7z3&*filter*=pe:2015&*filter* =ou:LEVEL-2;JLA7wl59oN3&displayProperty=SHORTNAME&outputIdScheme=ID& *aggregationType=STDDEV* Translated: give me the StDev for indicator dx:Ar2V5v1R7z3 for the period 2015 based on OU2 (population) values below 'JLA7wl59oN3: za South Africa'. At the moment this returns zero rows (i assume because dimension is only specified once) but when I switch from using 'ou filter' to 'ou dimension' i get values per ou2 that appear dramatically larger that the expected StDev value for the range. The same applies to the ou1 level (unusually large StDev values). E.g. indicator values for the first 3 months in 2015 are [20.6, 20.7, 20.7] while using aggregationType=STDDEV returns [45.5, 45.2, 47.9]... Kind Regards, Greg -- *Health Information Systems Program - South Africa* *- - - - - - - **- - - - - - - **- - - - - - - **- - - - - - - **- - - - - * Mobile : 073 246 2992 Landline: 021 554 3130 Fax: 086 733 8432 Skype: gregory_rowles
_______________________________________________ Mailing list: https://launchpad.net/~dhis2-devs Post to : dhis2-devs@lists.launchpad.net Unsubscribe : https://launchpad.net/~dhis2-devs More help : https://help.launchpad.net/ListHelp
