[R-sig-ME] reaped measures with unequal intervals

[Tom_Philippi at nps.gov]

Ahmad--

The unequal intervals part is rather simple.  Let Time be your variable
with values {6, 13,15}.  If you expect a linear trend over time within
subjects, then Time should be numeric; if not make Time a factor.  Equal
spacing of repeated measures would let you do something between these two
extremes: decompose trends over time into simple polynomial (linear,
quadratic, cubic, etc.) trends.  But, with only 3 dates, you wouldn't be
doing that anyway.

Everything else in the analyses depends on your experimental or sampling
design, response variable, etc., and the questions you are trying to
address with your analyses.  In my case of fixed islands, random sites
within islands, and balanced revisit dates (not evenly spaced because some
years are missing, but those years are missing for all sites), I need to
fit both an intercept and a temporal trend in my random effects: (1+Year |
Site).  With Year as numeric, I am fitting a single parameter linear trend
over time (estimating the variance among Plots in their linear temporal
trends).  With Year as a factor, I would be fitting nlevels(Year)
parameters for N : intercept (==Year 1) and 1 parameter for the difference
of each subsequent year (or (0+Year|Plot), which would give me
nlevels(Year) parameters, each directly interpretable as the (variance
across plots of the) mean for that Year).

I hope that this helps point you in the right direction.

Tom



                                                                           
             Ahmad Rabiee                                                  
             <AhmadR at sbscibus.                                             
             com.au>                                                    To 
             Sent by:                  "r-sig-mixed-models at r-project.org"  
             r-sig-mixed-model         <r-sig-mixed-models at r-project.org>  
             s-bounces at r-proje                                          cc 
             ct.org                                                        
                                                                   Subject 
                                       [R-sig-ME] reaped measures with     
             05/22/2012 04:51          unequal intervals                   
             AM GMT                                                        
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           




Hi

I have a repeated measures dataset with unequal intervals (spacing) - the
intervals are Day 6, Day 13 and Day 15.
Anyone knows how to handle this type of data?

Thanks
Ahmad


             [[alternative HTML version deleted]]

_______________________________________________
R-sig-mixed-models at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models