Price determination and inventory allocation based on spot and futures markets in future site channels for online advertising

A method that facilitates generating a price for a portion of an online advertisement, comprising:

• employing a processor executing computer executable instructions embodied on a computer readable storage medium to perform the following acts:
• evaluating at least one spot market and one future market by employing a probabilistic analysis that infers the starting ad price using a support vector machine that generates a computation of a probability distribution for different prices based on consideration of the data related to the future market and the spot market;
• automatically apportioning an inventory of a plurality of ad units based upon the evaluation for a publisher, wherein a subset of the inventory is apportioned to the spot market and a subset of the inventory is apportioned to the future market;
• and
• generating a set price for the inventory apportioned to the future market and a starting price for the inventory apportioned to the spot market based at least in part upon the evaluation.

One or more computer-storage hardware devices having computer-executable instructions embodied thereon that when executed by a computing device facilitates establishing a true market value for an online advertisement, comprising:

• receiving data related to at least one of a spot market or a future market;
• indentifying at least one sale within the spot market for an ad unit similar to an ad unit in the future market using a machine classifier that maps attributes of the at least one sale to attributes of the ad unit;
• automatically apportioning an inventory of a plurality of ad units based upon the data, wherein a subset of the inventory is apportioned to the spot market and a subset of the inventory is apportioned to the future market;
• dynamically establishing a set ad price for the ad unit in the future market based upon a sale price for the identified at least one ad sale within the spot market by employing a probabilistic analysis to infer the set ad price for the ad unit in the future market using a support vector machine that generates a probability distribution for different set ad prices based on ad sales within the spot market;
• and
• consummating transactions for the ad units in both the spot market and the future market based on the apportionment.