Tweet
BioMch-L
I have a group of 10 subjects (from a small population - elite lightweight
rowers) and for each subject I have kinematic data for 200 strokes.
I am attempting a multiple regression analysis on the data, I have one
outcome variable and between 6-8 input variables.
Can I use n=2000 (10x200) or do I need to have a different regression
equation for each person (n=200 by 10 regressions)?
If I choose either route do I need to use all 200 strokes per person or can
I reduce this number, digitising all 200 means 3.6 million points!!
Ross
---------------------------------------------------------------
To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l
---------------------------------------------------------------
I have a group of 10 subjects (from a small population - elite lightweight
rowers) and for each subject I have kinematic data for 200 strokes.
I am attempting a multiple regression analysis on the data, I have one
outcome variable and between 6-8 input variables.
Can I use n=2000 (10x200) or do I need to have a different regression
equation for each person (n=200 by 10 regressions)?
If I choose either route do I need to use all 200 strokes per person or can
I reduce this number, digitising all 200 means 3.6 million points!!
Ross
---------------------------------------------------------------
To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l
---------------------------------------------------------------