[R-sig-ME] Need help to find multifactors influence‏

[Muhammad Arshad]

Dear friends,

1-I have B.I value (Dependent variable).
2-I have land cover types in the form of % area against each B.I (that was
clipped through buffer for each B.I ). Buffer was put by considering B.I
behavior that can be influence by land type.
3- I want to see the influence of land type on B.I.
4- Suppose B.I value varies with the change in (area) one or more types of
land cover. This change may be linear up to some extent but can also
increase with the combination of land types. Land types can also differ
from one buffer to other buffer.
5- I want to develop a Predication Model\Regression Equation to predict B.I
value with the combined influence of different land use types.
6- I am confused which Regression model (Linear, Logistic, and Poisson,
Exponential etc...)  Should be adopted to develop this type of relationship
where factors are varying.
7- Total study map consists of 77 land cover types and each buffer gets
some land type.

Please see the attached data having values extracted for each land type
(a1~a77 are land type names against each BI).

I already tried to solve it with different methods. Even I made groups in
term of B.I & also in term of land cover types. But still failed to
find correlation with reasonable factors.

I will happy for your informative response to solve my this problem. Please
let me know if my question is not clear.

Warm Regards,
Malik\Arshad