Dynamic Pricing models are used to dynamically change prices, either at relatively high frequency, or on-demand, i.e. each time a customer queries about the price for a given product or service. The parameters in dynamic pricing can be many, including:

• costs
• capacity
• stock
• customer's willingness-to-pay
• customer's previous purchase patterns
• margin expectations
• competitor prices
• seasonality
• customer behaviour on e.g. a website (are they first time visitors or have they queried the price before)
• price elasticity
• and much, much more.

Price elasticity of demand has been in the economic text books for many decades now and while the classical demand curves with their price elasticity assumptions have rightfully been criticised for being too far away from reality, there is still room for the use of price elasticity in dynamic pricing models, in particular in industries and businesses with high transaction volumes.

Having statistically significant numbers for price elasticity allows for modelling what happens when price changes in the model. Similarly, other elasticities for any of the other parameters, such as competitor price changes, can be used to make dynamic price changes when the model predicts it is necessary. Or to stay put despite competitive price changes, because the model "knows" that the volumes sold for this specific product is not impacted by competitors' moves. 

The operative phrase above is "statistically significant". In other words, if you have a lot of historical transactions to base the elasticity parameters on in the model, then it is often a good thing to use price elasticity in dynamic pricing. If on the other hand you have only few data points, then modelling based on historical data is not useful.