Hello,
How to remove unimportant trend - cycle from time series data? (sensor data)
I have tried to use seasonality corrections, but the results which i have received are not correct(Objective value: NaN)
Cheers
We currently don't offer a seasonal decomposition in KNIME. For so long I would recomment using R for this, here is a nice description for STD in R: http://www.r-bloggers.com/seasonal-trend-decomposition-in-r/
A naive approach in KNIME (if you don't want to use R) is to calculate the mean value for each season (e.g. each day/ each hour) and remove it from matching data point.
I hope this helped.
Best, Iris
I have sent here the sample of my data, but since 4 days it hasnt appear here as a comment? why is that?
Dear Malvina,
I don't see any posts containing data in this thread. Maybe you are refering to your other thread, there I just answered.
Best, Iris
Yes i do, i am sorry for that! Thank You so much for the tipp for previous thread
Greetings