Thanks for quick responses.
My dataset consists of 213 independent variables (disease related costs) and
the depedent variable is the total medical + pharmacy cost of the member.
As you had suggested, I tried to use hier.part function in R, but the
current implementation does not seem to allow
Rupendra Chulyadyo wrote:
Hello all,
I want to assign relative score to the predictor variables on the basis of
its influence on the dependent variable. But I could not find any standard
statistical approach appropriate for this purpose.
Please suggest the possible approaches.
Thanks in
I suggest using permutation on each predictor and see how much the
accuracy drops, no matter what modeling approach you used.
HTH,
weiwei
On 1/17/07, Rupendra Chulyadyo [EMAIL PROTECTED] wrote:
Hello all,
I want to assign relative score to the predictor variables on the basis of
its
Hello all,
I want to assign relative score to the predictor variables on the basis of
its influence on the dependent variable. But I could not find any standard
statistical approach appropriate for this purpose.
Please suggest the possible approaches.
Thanks in advance,
Rupendra Chulyadyo
Rupendra,
depending on the nature of your data (which you haven't mentioned),
you might try hierarchical partitioning, as found in the hier.part
package on CRAN.
Cheers
Andrew
On Wed, Jan 17, 2007 at 11:07:18AM +0545, Rupendra Chulyadyo wrote:
Hello all,
I want to assign relative score to
Before you do that, you might try reading this paper:
Bring, J. 1995. Variable importance by partitioning R^2. Quality and
Quantity 29:173-189.
Cheers,
Simon.
Andrew Robinson wrote:
Rupendra,
depending on the nature of your data (which you haven't mentioned),
you might try hierarchical