Logistic equation

ravenwolf at my-dejanews.com ravenwolf at my-dejanews.com
Sat Jun 20 11:11:31 EST 1998


In article <35894CF3.4946 at mail.unimo.it>,
  Albertino Bigiani <bigiani at mail.unimo.it> wrote:
>
> Can anybody tell me what "logistic" stands for in the so-called
> "logistic equation" used to fit dose-response data?
>
> Thanks
>
> Albertino Bigiani
> When you work with some kind of variables that in the interest of you research

you prefer to categorize it in a "yes" "no" type, (for example) instead of
use a continous variables you can run those data under a logistic equation,
For example:

Group 1      10 subjets      dosis 1--------------> response 3+     7-
Group 2      10 subjets      dosis 2--------------> response 5+     5-
group 3      10 subjets      dosis 3--------------> response 4+     6-
Group 4      10 subjest      dosis 4--------------> response 8+     2-

so qhe you enter data you will set them like this:
Dosis     response         dosis         response
1           1                2               1
1           1                2               1
1           1                2               1
1           0                2               1
1           0                2               1
1           0                2               0
1           0                2               0
1           0                2               0
1           0                2               0
1           0                2               0

etc, so whe you run a logistic analysis you will analize the proportion of
"1" and "0" for each dose like they be numeric variables. Like this: dosis
1--------> 0.3 dosis 2--------> 0.5 dosis 3--------> 0.4 dosis 4--------> 0.8
Generally statistics program can run regretion logistic analisys without
problem qith the original set of data, but in other programs you must enter
the proppiortions just calculated and you must indicate the program the
number of subjet or observation used for each proportion, in other cases you
will be asked  to use the same number of observatioon in each level, but
generally statistics programs could made the necesary corretion for unequal
number of observation . Luck

-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/   Now offering spam-free web-based newsreading



More information about the Neur-sci mailing list