Minute-by-minute vs summary data

[alexperto]

Hi guys, 

We're integrating step-time-series data in our system for a medical study and we have detected some discrepancies comparing minute-by-minute vs the daily data in almost all our users.

First of all, I don't fully understand how do you 'generate' the time series data, is it calculated at the end of the day? is it saved in the device?

Anyways, I'm using an script to sum up the intraday values to see if they match with the daily ones and  this a result of my data from the past month:

date: 09/28/2015 => daily summary: 10942 , intraday sum: 10942 

date: 09/29/2015 => daily summary: 6015 , intraday sum: 6015
date: 09/30/2015 => daily summary: 11011 , intraday sum: 11011
date: 10/01/2015 => daily summary: 4734 , intraday sum: 4734
date: 10/02/2015 => daily summary: 6304 , intraday sum: 6304
date: 10/03/2015 => daily summary: 16180 , intraday sum: 16180
date: 10/04/2015 => daily summary: 5042 , intraday sum: 5042
date: 10/05/2015 => daily summary: 18 , intraday sum: 18
date: 10/06/2015 => daily summary: 12180 , intraday sum: 12180
date: 10/07/2015 => daily summary: 5576 , intraday sum: 5576
date: 10/08/2015 => daily summary: 4732 , intraday sum: 4732
date: 10/09/2015 => daily summary: 13133 , intraday sum: 13133
date: 10/10/2015 => daily summary: 11035 , intraday sum: 11035
date: 10/11/2015 => daily summary: 5363 , intraday sum: 5363
* date: 10/12/2015 => daily summary: 5109 , intraday sum: 3470
date: 10/13/2015 => daily summary: 11396 , intraday sum: 11396
date: 10/14/2015 => daily summary: 5926 , intraday sum: 5926
date: 10/15/2015 => daily summary: 5619 , intraday sum: 5619
date: 10/16/2015 => daily summary: 12235 , intraday sum: 12235
date: 10/17/2015 => daily summary: 9693 , intraday sum: 9693
date: 10/18/2015 => daily summary: 3467 , intraday sum: 3467
date: 10/19/2015 => daily summary: 8111 , intraday sum: 8111
* date: 10/20/2015 => daily summary: 7354 , intraday sum: 7341
* date: 10/21/2015 => daily summary: 6191 , intraday sum: 0
date: 10/22/2015 => daily summary: 6202 , intraday sum: 6202
date: 10/23/2015 => daily summary: 5469 , intraday sum: 5469
date: 10/24/2015 => daily summary: 6942 , intraday sum: 6942
date: 10/25/2015 => daily summary: 7680 , intraday sum: 7680
date: 10/26/2015 => daily summary: 3831 , intraday sum: 3831
date: 10/27/2015 => daily summary: 3333 , intraday sum: 3333
date: 10/28/2015 => daily summary: 1548 , intraday sum: 1548

Notice that for 10/12/2015, 10/20/2015 & 10/21/2015 we have inconsistencies.

The weirdest case for me is on 10/21/2015 where I have zero intraday data.

Here are more details about these cases:

10/12/2015

10/20/2015

10/21/2015

Hope you can help me to understand what's going on!

Thank you so much

[stephmsherman]

Hello,

We are having the same issue with our data - there is a major discrepancy between the minute-by-minute and the daily data. Did you ever discover the source of the problem?

Thanks!

[AndrewFitbit · Accepted solution]

@stephmsherman This post from Jeremiah should answer your question: Intraday values