Skip to main content
12 events
when toggle format what by license comment
Dec 22, 2017 at 16:25 answer added Russ Harris timeline score: 1
Apr 30, 2012 at 12:42 vote accept jermeyz
Apr 28, 2012 at 2:53 answer added Peter Ellis timeline score: 4
S Apr 27, 2012 at 21:19 history suggested Emre CC BY-SA 3.0
copy-editing
Apr 27, 2012 at 20:02 comment added PhD I may be wrong, but did you try simple correlation to see if it even makes sense? Correlating sales with new design for that time period vs that of the old one? Intuitively this makes sense for a cursory analysis before diving into details, but I'll leave it up to the experts to comment on this
Apr 27, 2012 at 19:58 comment added IrishStat f you actually post your data I will try and help as I have had extensive experience in modelling daily data taking into account day-of-the-week , month of the year, holiday effects , particular days of the month effects , level shift effects and local time trends. What needs to be done is to empirically detect a substantive change which could be assignable to a number of possible "causes".
Apr 27, 2012 at 18:53 answer added pgericson timeline score: 3
Apr 27, 2012 at 18:19 review Suggested edits
S Apr 27, 2012 at 21:19
Apr 27, 2012 at 16:42 history edited Peter Flom CC BY-SA 3.0
Fixed title to reflect topic, added time series tag
Apr 27, 2012 at 15:36 comment added jermeyz We have their revenue numbers in dollars of products sold for about the last two years on a daily basis. We are trying to determine if a recent drop in revenue after a site relaunch can be attributed to the redesigned web site functionality or just a normal variance.
Apr 27, 2012 at 15:27 comment added Ken Can you be a little more specific in terms of data you have?
Apr 27, 2012 at 15:23 history asked jermeyz CC BY-SA 3.0