| |
 |
|
|
Science Forum Index » Statistics - Math Forum » simple repeated measures test of acceleration
Page 1 of 1
|
| Author |
Message |
| Guest |
 Posted: Wed Jan 24, 2007 6:59 pm |
|
|
|
|
I think this is simple but I can't think of how to begin.
I have between 8 and 15 time measurements in a series. For example the
following values measured in seconds; 0.23, 0.35, 0.24, 0.44, 0.38,
0.40. I'm trying to figure out if they are generally increasing or
decreasing in value, ie accelerating, decelerating, or holding steady.
Actually, I'd like to plot it on excel and just look at the averaged
slope to see if it is going up or down. But I can't figure it out.
Any ideas? |
|
|
| Back to top |
|
| Ray Koopman |
| Posted: Wed Jan 24, 2007 9:22 pm |
|
|
|
Guest
|
On Jan 24, 2:59 pm, contact.c...@gmail.com wrote:
Quote: I think this is simple but I can't think of how to begin.
I have between 8 and 15 time measurements in a series. For example the
following values measured in seconds; 0.23, 0.35, 0.24, 0.44, 0.38,
0.40. I'm trying to figure out if they are generally increasing or
decreasing in value, ie accelerating, decelerating, or holding steady.
Actually, I'd like to plot it on excel and just look at the averaged
slope to see if it is going up or down. But I can't figure it out.
Any ideas?
Try the Mann-Kendall nonparametric trend test. Look at each of
the n(n-1)/2 pairs of measurements: let P be the number of pairs
in which the second measure is greater than the first measure,
indicating an increasing trend; and let Q be the number of pairs in
which the second measure is less than the first measure, indicating
a decreasing trend. The descriptive statistic is (P-Q)/(n(n-1)/2),
which will be between +1 and -1. To test its significance, refer
(P-Q)/sqrt[n(n-1)(2n+5)/18] to the standard normal distribution. |
|
|
| Back to top |
|
| |
|
Page 1 of 1
All times are GMT - 5 Hours
The time now is Sun Jul 20, 2008 2:11 am
|
|
