This dissertation explores several new directions in both theoretical (chapter 1) and empirical (chapters 2 and 3) macroeconomic models. The first chapter offers a theoretical explanation of why agents save when asset prices fall: lower asset prices bring individuals closer, in real terms, to a potentially binding debt constraint. To avoid becoming debt constraint agents thus increase their buffer wealth. The second chapter bridges the gap between theory and empirics by solving and estimating a simple DSGE model of stochastic growth. It then applies the model to estimates of monthly U.S. GDP and nowcasts of quarterly GDP. The final chapter is purely empirical and looks at the distinction between idiosyncratic and persistent errors in observables when estimating factor models. This distinction is particularly useful for identifying measurement error.