This research project aims to understand how reference-dependent behaviours of consumers affect the aggregate level market response as well as how the firm should make optimal pricing decisions. The project investigates this question from two separate perspectives. First, the dynamic pricing literature with reference effect commonly imposes assumptions about consumer behaviour directly on the market response. While this may be convenient for analytical purposes, it may not accurately reflect the real market response. We propose to bridge this gap and examine the analytical and empirical consequences when one builds up the reference dependent demand function from the consumer behaviour model. Second, we analyse the problem of determining the optimal prices in more realistic settings such as reference prices can evolve stochastically or the firm needs to manage the pricing decisions for multiple products.