We have some measurements on the eye that we suspect have some kind of
diurnal pattern.
We only have the measurements during 'office hours' and they were only
collected quite sparsely (5 measures at 9, 11, 13, 15 and 17 hrs) but at
least we have measurements of several subjects (n=61).
We suspect there is a pattern in the sequence of measurements during
these 5 time points. We could simply reduce each time series to a
sequence, e.g. + - + - where, for example, the measure would increases,
decrease, increase, decrease. Of course there are some where the
measurements are ++++ (increase, increase, increase, and increase) etc
etc.. I was thinking along the lines of testing whether the observed
sequences have an 'average' pattern, or that the observed patterns are
beyond what we might expect by chance - given of course that we are more
likely to see +- than ++. We could equally normalize these sequences
such that they are above or below an average reading from the day.
Any thoughts, pointers or help would be most gratefully received.
Dr. David Crabb
Department of Optometry and Visual Science, City University, Northampton
Square, London EC1V OHB
Tel: 44 207 040 0191 [log in to unmask]
http://www.city.ac.uk/optometry/about/staff/crabb.html
<http://www.city.ac.uk/optometry/about/staff/crabb.html>
<http://www.city.ac.uk/optometry/html/david_crabb.html>
|