Systems and methods for optimal bidding in a business to business environment
The present invention relates to systems and methods for optimizing bidding in a business-to-business environment. Initially the observed outcomes for n deals are received, and the belief parameters for these n deals are calculated. The Bayes-greedy price is then calculated and presented to a buyer. The buyer's response is collected and an optimal variance parameter based on the buyer's response is generated. The belief parameters for these n+1 deals are also updated. This process may be repeated for additional deals.
1. A method for reducing computations of non-normal data with a processor, comprising:
receiving features of, and observed outcomes for, n deals;
calculating, via a processor, belief parameters for the n deals, the belief parameters being stored in memory associated with the processor, the belief parameters being a normal distribution of the outcomes based on the features;
when a new deal is received:
calculating, via the processor, a posterior distribution representing regression coefficients of the observed outcomes in conjunction with the new deal, the posterior distribution being a non-normal distribution;
replacing, within the memory associated with the processor, the posterior distribution with an approximated distribution of the posterior distribution, the approximated distribution having a normal distribution; and
calculating, via the processor, a Bayes-greedy price for future deals using the approximated distribution and the belief parameters for the n deals,
the Bayes-greedy price calculated as: p k−1 =p k −α k ∇ p k R(p k :x,β);
outputting the Bayes-greedy price to a display;
adding the new deal to the observed outcomes; and
recursively updating the belief parameters based on the new deal.
2. The method of claim 1 , wherein the belief parameters for then deals is calculated by:
θ
′
=
θ
-
v
(
Y
-
1
2
)
-
x
T
θ
v
+
x
T
Σ
x
Σ
x
.
and
Σ
′
=
Σ
-
Σ
xx
T
Σ
v
-
x
T
Σ
x
.
3. The method of claim 1 , further comprising calculating an optimal variance parameter based on the buyer's response.
4. The method of claim 3 , wherein the optimal variance is calculated by:
=
-
α
k
∇
v
k
D
KL
(
Q
❘
P
)
5. The method of claim 3 , further comprising calculating belief parameters for the n+1 deals.
6. The method of claim 5 , further comprising repeating the method for deal n+2.
7. The method of claim 5 , further comprising repeating the method for additional deals.