Re: Comparing trajectories
- From: EinsteinHelpMe <mandadopor-scimath@xxxxxxxxx>
- Date: Fri, 07 Sep 2007 08:14:50 -0700
On Sep 7, 8:13 am, isabelle...@xxxxxxxxxxx wrote:
Dear Forum,
I am working on a applied project whose goal is, for each subject, to
compare two trajectories. The trajectories represent functional
decline in the last year of life. For each subject, one trajectory is
drawn, the other one consists of measurements of functional status
across time.
I am looking for a method/piece of software that can help me
compare those trajectories. I would like to be able to consider both
curve and scale relative to the potential range.
The sample size is N=30. I was told that the Kolmogorov Smirnov test
can be used. I am skeptical because there are so few subjects.
Many thanks for your input.
Hi Isabelle:
If sampling is uniform I guess you could stack all the measurements in
a vector (30X1 vector) and compare the obtained vector. A simple
measure of how close they are could be |a-b| where |.| might be the
standard norm or even some other norm like the Frobenius norm (http://
mathworld.wolfram.com/FrobeniusNorm.html). The lower the norm, the
closer a and b are. I don't know the Kolmogorov Smirnov test so I
can't tell you about it.
Hope this helps,
Sergio
sergiodlc -AT- gmail.com
.
- References:
- Comparing trajectories
- From: isabellesup
- Comparing trajectories
- Prev by Date: Re: Brain teaser
- Next by Date: Some basic problems in Stochastic Calculus
- Previous by thread: Comparing trajectories
- Next by thread: Re: Comparing trajectories
- Index(es):
Relevant Pages
|
