LAUSR

Abstract Using a discrete-time asset-pricing model, I specify the economic rationale for a rich array of price dynamics. Two boundedly-rational investors with different risk preferences trade periodically, where excess supply is cleared by a tâtonnement process. Cast at the core of asset-pricing modeling, this structure allows me to explore price discovery intra-periodically, and over time. If dividends are observable, the price converges to Merton's ICAPM, but if investors rely on past realizations, momentum and reversal patterns emerge, which might escalate to bubbles and crashes. The model features increasing volume but declining liquidity during positive bubbles, and lowest liquidity after negative bubbles.