How to model/test some financial data

Hi,

I was hoping some one could please give me some guidance. I have
datasets that contain details of the financial performance of
different groups of individuals as well as another for those not in
any group. My expectation was to see a difference in the performance
of those individuals in groups relative to those not in a group due to
the differing levels of support they receive. I measure performance
in terms of their ability to get a loan at the lowest possible rate.
The data is from a peer-to-peer lending site where individuals lend to
other individuals through a reverse auction type process i.e. lenders
bid down the rate at which they are willing to lend to the individual
looking for the loan.

I have one independent variable which is group membership. The
dependant variables are percentage of loan funded (by other
individuals), the lender rate and the rate saving (i.e. difference
between the max rate the individual was willing to pay and the rate at
which the prosposed loan currently stands). I have also included some
other variables that might influence the result i.e. group type (i.e.
whether the group is open or invite only) and individual credit
rating.

My problem is that it is many years since I studied mathematics and
I'm struggling to see a way forward. I wish to confirm or reject a
number of hypotheses:

P1 - Borrowers that are members of groups are more likely to get
funded that those that are not.
P2 - Borrowers that are members of groups are more likely to secure a
lower interest rate than those that are not.
P3 - Borrowers that are members of groups are more likely to secure a
greater rate saving than those that are not.

While I can see some trends in the data, I'm not so sure they are
statistically significant. I was considering doing a t-test between
the average for the group individuals and the average for the non-
group individuals. However, I recall that regression testing may also
be of benefit in deciding whether or not various factors are
significant in influencing the dependent variables. Any help or
advice as to how I should proceed would be much appreciated.

.

Relevant Pages

  • Re: How to model/test some financial data
    ... P2 - Borrowers that are members of groups are more likely to secure a ... group B has 60% college graduates and a loan default rate of 2%, ...
    (sci.stat.math)
  • help with criteria
    ... somewhat similar to your subquery idea, ... You could try first creating a query that joins the ... members and loan info for only the loan types that you ...
    (microsoft.public.access.queries)
  • Re: My opinion of the BN Change Group
    ... a loan was also made to Steve Thompson 'Official' ... Nor am I suggesting that the expenditure in the 2005 accounts labelled ... Are BN members to take your above statement Duncan that you think the above ...
    (uk.rec.naturist)
  • Re: My opinion of the BN Change Group
    ... a loan was also made to Steve Thompson 'Official' ... Are BN members to take your above statement Duncan that you think the above ... to post an opinion that differs from theirs or seems to support BN, ...
    (uk.rec.naturist)
  • Eagle Surface Grinder
    ... I would like the loan, or copy of a manual or information on an Eagle ... Surface Grinder for one of our members ...
    (uk.rec.models.engineering)