I have a question re time series analysis: I have multiple t Dear All, I have a question re time series analysis: I have multiple time series of litter layer temperature and want to compare the series to show that they are/are not different from each other in terms of mean and/or amplitude.
In the literature, one can see that others often have used repeated-measures ANOVA. But I do not think that this is a good approach, certainly not for long time series with thousands of data points. I have the feeling that I should decompose the time series into their trend, daily and random components and then look for differences in there, but I am not really sure whether this is the right way. Another option would be to develop Autoregressive Integrated Moving Average (ARIMA) Models (or similar) and then compare the model parameters for differences. But again I am not totally sure this is the way to go. Somehow I have the feeling that this problem should occur more often, but I cannot find much online or in the ECOLOG-Archive. Any suggestions and insights would be greatly appreciated. Best regards, Michael Drescher
