Trixology
WeatherCat => WeatherCat General Discussion => Topic started by: vetenskapsman on November 13, 2016, 08:02:33 PM
-
I've a question about how Averages are calculated in the Year Over Year Stats. It seems to treat the current month as a full month and thereby gives too much weight to the early days of the current month. For instance, I've only 3 years of data for November. Using one example of my reported stats for Wind Run as of Nov. 13 ...
Wind Run: 2016 - 280
2015 - 910
2014 - 1140
Average: 777 which is (280+910+1140)/3
But we are only 13 days into November. A closer Average would be (280+910+1140)/(2+(13/30) = 958
Currently the average is furthest off on the 1st day of the month and then is closer to the correct average on the last day of the month. Maybe there is a good reason for it to be this way but I thought I'd ask?
best, -carl
-
"You can prove anything with statistics (averages)!", A. Einstein. That's why I don't publish anything but the actual high/low/etc. for whatever time frame. [lol]
-
Dear Carl, X-Air, and WeatherCat "statisticians" . . . .
I've a question about how Averages are calculated in the Year Over Year Stats. It seems to treat the current month as a full month and thereby gives too much weight to the early days of the current month.
Your math looks correct to me, but as you say perhaps there is some rationale for this way of calculating. If Stu (Stuart Ball, WeatherCat developer) doesn't chime in with an answer, perhaps you should contact WeatherCat technical support and ask.
Cheers, Edouard