Timeline for Data has two trends; how to extract independent trendlines?
Current License: CC BY-SA 3.0
17 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 9, 2023 at 21:04 | comment | added | Markus | Is your solution for price data and technical analysis? If not, what is the problem domain? | |
| Aug 15, 2012 at 5:53 | history | tweeted | twitter.com/#!/StackStats/status/235615232469397504 | ||
| Aug 14, 2012 at 12:27 | answer | added | Steven | timeline score: 17 | |
| Jul 31, 2012 at 11:14 | vote | accept | jonathanbsyd | ||
| Jul 29, 2012 at 13:48 | comment | added | jonathanbsyd | No the client can't seperate the records with the available information. The 2012 data will improve as they clean up their recent data, we are getting the next iteration of this data in the coming days with the 2012 data fixed and brought up to date. While I can't disclose what this data is of, I can say it is definitely what the client wants, to find out how consumption will move in the coming years. | |
| Jul 26, 2012 at 16:43 | history | edited | user88 | CC BY-SA 3.0 |
edited tags; edited title
|
| Jul 26, 2012 at 16:24 | answer | added | whuber♦ | timeline score: 30 | |
| Jul 26, 2012 at 15:08 | answer | added | Wayne | timeline score: 2 | |
| Jul 26, 2012 at 15:00 | comment | added | Wayne | Just to make sure: your client doesn't have any additional data that would indicate which measurements come from which population? This is 100% of the data that you or your client have or can find. Also, 2012 looks like either your data collection fell apart or one or both of your systems fell through the floor. Makes me wonder if trend lines up to that point matter much. | |
| Jul 26, 2012 at 12:49 | history | migrated | from stackoverflow.com (revisions) | ||
| Jul 25, 2012 at 19:50 | comment | added | Marcin | Could you crudely filter the data to isolate the two trendlines, then use those trendlines on the whole dataset to identify points on the trend less crudely, and separate the dataset on that basis? | |
| Jul 25, 2012 at 19:46 | answer | added | user1149913 | timeline score: 31 | |
| Jul 20, 2012 at 1:55 | comment | added | jonathanbsyd | Unfortunately this isn't effect data but usage data, and clearly usage from two separate systems mixed up into the same data set. I want to be able to describe the two usage patterns, but I can't go back and recollect data as this represents about 6 years worth of information collected by a client. | |
| Jul 19, 2012 at 23:45 | comment | added | Matt Parker | I don't have any personal experience with it, but I think statsmodels would be worth checking out. Statistically, a linear regression with an interaction for group would be adequate (unless you're saying you have ungrouped data, in which case that's a bit hairier...) | |
| Jul 19, 2012 at 14:34 | comment | added | Thomas Jungblut | Maybe you can compute pair-wise slopes and group them to two "slope-clusters". However this will fail if you have two parallel trends. | |
| Jul 19, 2012 at 4:15 | comment | added | jonathanbsyd | Best answer I have so far is to print this on graph paper and use a pencil and ruler and calculator ... | |
| Jul 19, 2012 at 3:28 | history | asked | jonathanbsyd | CC BY-SA 3.0 |