d******3
发帖数: 93 | 1 e.g. 1000 observations, 10 variables x1, x2, x3, ... x10. x1 is continuous,
x2-x10 could be continuous or categorical (e.g. age, gender, race...)
now I want to divide these 1000 obs into several groups (e.g. 3 groups), and
to maximize the difference in x1 across these 3 groups while minimize the
differences in x2-x10 across groups.
thanks | D******n
发帖数: 2836 | 2 multiple objective optimization doesnt necesarily have a solution.
Easiest way is to combine multiple objectives with weights
S_single=w1S1+w2*S2+....wkSk
,
and
【在 d******3 的大作中提到】
: e.g. 1000 observations, 10 variables x1, x2, x3, ... x10. x1 is continuous,
: x2-x10 could be continuous or categorical (e.g. age, gender, race...)
: now I want to divide these 1000 obs into several groups (e.g. 3 groups), and
: to maximize the difference in x1 across these 3 groups while minimize the
: differences in x2-x10 across groups.
: thanks
| l*********s
发帖数: 5409 | 3 The question is interesting but a little vague. Are these groups have to
have roughly same number of observations? If not, what is the formal
definition of difference? | d******3
发帖数: 93 | 4 thanks a lot for your replies
I agree that this question is a little vague
What if there is no restrictions on how many number of observations, I think
it OK to have 4 groups with 100, 200, 300, 400 each. | d******3
发帖数: 93 | 5 this is really interesting. thanks!
what if we have categorical variables, e.g. x3 is gender, male or female, x4
is race, black or not, and x1, x2... are continuous variables such as age,
income...
【在 D******n 的大作中提到】
: multiple objective optimization doesnt necesarily have a solution.
: Easiest way is to combine multiple objectives with weights
: S_single=w1S1+w2*S2+....wkSk
:
: ,
: and
|
|
