Time Series and Macroeconomics: Studies in Demography and Monetary Policy

Sammanfattning: Chapter 1 provides a brief introduction to the topics of the thesis and summarises the main results.Chapter 2 studies the econometric properties of the Taylor (1993) rule when applied to U.S., Australian and Swedish data in order to judge its empirical relevance. Unit root tests indicate that the variables commonly used in such modelling are likely to be integrated of order one or near integrated. Given these time series properties, cointegration becomes a necessary condition both for consistent estimation of the parameters of the model and compatibility between the model and the data. Tests find little support for cointegration and, together with an out-of-sample forecast exercise, suggest that we should have serious doubts about the Taylor rule as a reasonable description of how monetary policy is conducted in the countries considered in this study.Chapter 3 investigates the relationship between age structure and GDP in 20 OECD countries using annual data between 1970 and 1999. Using new methodology, the relationship between the variables can be formulated in levels despite the presence of unit roots in the time series. Applying two panel cointegration tests proposed by Pedroni (1999), support is found for a long run relationship between GDP and the number of people in five different age groups. Coefficient estimates from panel regressions support effects predicted by the life cycle hypothesis and human capital theory.Chapter 4 investigates the small sample performance of four well-known cointegration tests when a system has been misspecified by leaving out one relevant variable from a system with one cointegrating vector. In a Monte Carlo study, the size distortions of the Augmented Engle-Granger (Engle and Granger, 1987), Johansen’s (1988) maximum eigenvalue, Johansen’s (1991) trace and the Boswijk (1989) Wald tests are examined. The Johansen trace test adjusted by a finite sample correction is found to have the most robust performance when lag length in the test equations is chosen according to traditional information criteria.Chapter 5 tests for the presence of unit roots in four U.S. macroeconomic time series using panel unit root tests. This is done by introducing a new panel setting which allows the researcher to test several hypotheses for one country instead of one hypothesis for several countries, which has been the standard approach in previous work. The Im, Pesaran and Shin (2003) test, the Multivariate Augmented Dickey-Fuller test (Taylor and Sarno, 1998) and the Johansen (1988) likelihood ratio test are applied to unemployment, the real exchange rate, the nominal interest rate and inflation. The three tests all have ways of controlling the obvious cross-sectional dependence in the panel. Using monthly data from 1960 to 2002 there is evidence that all time series are generated by stationary processes.