Survey_BC_2012

Abstract

In order to evaluate the performance of the long-run risks model, we follow BY and use data on US nondurables and services consumption from the Bureau of Economic Analysis. We consider both a long-run annual series over the period 1930-2006, and a postwar quarterly US series over the period 1947:2-2007:3. Table II reports basic annual moments for US data over the period 1930-2006, and the annual population moments implied by the BY and BKY calibrations of the long-run risks model with relative risk aversion = 10. The population moments are calculated from a single simulation run over 1.2 million months or 100,000 years. Table III repeats this exercise for quarterly postwar US data and the quarterly population moments from the model. We look at Öve variables: the changes in log consumption and dividends, log stock return, log riskfree interest rate, and log price-dividend ratio. All variables are measured in real terms. For each variable, we report the mean, standard deviation, and Örst-order autocorrelation.