R help - Jun 2011 - Time-series analysis with treatment effects - statistical approach

Hello all R listers,
I'm struggling to select an appropriate statistical method for my data set.
I have collected soil moisture measurements every hour for 2 years. There
are 75 sensors taking these automated measurements, spread evenly across 4
treatments and a control. I'm not interested in being able to predict soil
future soil moisture trends, but rather in knowing whether the treatment
affected soil moisture response overall.  In particular, it would be
interesting to inspect treatment related response within defined periods.
For example, a visual inspection of my data suggests that soil moisture is
equivalent across treatments during wet winter months, but during the dry
summer months, a treatment effect appears. 

Any help on this topic would be very appreciated. I've looked far and wide
through academic literature for similar experimental designs, but have not
had any success as yet.

Cheers,
Justin

Dr. Justin Morgenroth
New Zealand School of Forestry
Christchurch, New Zealand




--
View this message in context:
http://r.789695.n4.nabble.com/Time-series-analysis-with-treatment-effects-statistical-approach-tp3615856p3615856.html
Sent from the R help mailing list archive at Nabble.com.

> Subject: [R] Time-series analysis with treatment effects - statistical
approach
> 
> Hello all R listers,
> I'm struggling to select an appropriate statistical method for my data
set.
> I have collected soil moisture measurements every hour for 2 years. There
> are 75 sensors taking these automated measurements, spread evenly across 4
> treatments and a control. I'm not interested in being able to predict
soil
> future soil moisture trends, but rather in knowing whether the treatment
> affected soil moisture response overall.  In particular, it would be
> interesting to inspect treatment related response within defined periods.
> For example, a visual inspection of my data suggests that soil moisture is
> equivalent across treatments during wet winter months, but during the dry
> summer months, a treatment effect appears. 
> 
> Any help on this topic would be very appreciated. I've looked far and
wide
> through academic literature for similar experimental designs, but have not
> had any success as yet.

Can you post your data? This makes it more interesting for potential responders
and makes it more likely you get a relevant answer or observation. 

I guess the stats people would just suggest something like anova and
you could just compare moisture means among the treatments and controls
but presumably things are not stationary and it may be easy to get confusing
results. 

http://en.wikipedia.org/wiki/Analysis_of_variance



Usually there is some analysis plan made up a priori I would imagine or at least
there is prior literature in your field. I guess citeseer or google
scholar/books may
be useful if you have some key words or author names. The data you have
sounds to an engineer just like any other discrete time signal and maybe digital
signal processing literature would be of interest. Presumably you have some 
models to test, and more details depend on the specifics of your competing
hypotheses. Kalman filtering maybe? 


> 
> Cheers,
> Justin
> 
> Dr. Justin Morgenroth
> New Zealand School of Forestry
> Christchurch, New Zealand
> 
> 
> 
> 
> --
> View this message in context:
http://r.789695.n4.nabble.com/Time-series-analysis-with-treatment-effects-statistical-approach-tp3615856p3615856.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
 		 	   		  
	[[alternative HTML version deleted]]