Assuming that I have the history of the demand as a function of price $p \in [p_{min},p_{max}]$. I need a model better than the linear regression to predict the demand and its elasticity. The model should be able to select the most relevant historical informations (the minimum amount possible like in the compressive sampling theory). I will exploit this model to choose the prices that maximizes the revenue and explore prices that doesn't belong to $[p_{min},p_{max}]$ (bandit theory). Do you know any complete demand prediction model that fulfills those requirements?
Thanks in advance