Three essays on applied time series econometrics

This dissertation consists of three independent essays in applied econometrics.;Essay Two: Convergence in the Canadian Provinces: Evidence using Unemployment Rate. Quarterly time series data from Canada and the Canadian provinces for the period 1976:1--2005:3 are examined to determine if the unemployment rates in the Canadian provinces are converging to the national rate of unemployment. The paper checks for existence of stochastic convergence using recent unit root statistics, see Perron and Rodriguez (2003) and Rodriguez (2007); furthermore, it verifies the existence of beta-convergence using recently proposed methods by Vogelsang (1998), Perron and Yabu (2005, 2006), and Bai and Perron (1998, 2003).;Results from different unit root tests, without and with structural breaks, confirm that stochastic convergence holds in all provinces except British Columbia. Next, the existence of beta-convergence is tested using different approaches to estimate the trend function. The results show that deterministic convergence holds and the unemployment rates of the Canadian provinces are converging to the unemployment rate of Canada. This conclusion is stronger when multiple breaks are allowed in the trend function, that is, using the approach of Bai and Perron (1998, 2003).;Essay Three: Persistence in Unemployment: Evidence from Canada and the Canadian Provinces. This paper addresses the persistence of unemployment in Canada. To test this issue, we use quarterly unemployment rate data from Canada and the Canadian provinces from 1976--2005. First, we utilize different unit root tests, without and with structural breaks, and confidence intervals for the autoregressive coefficients, to test the null hypothesis of unit root. The results of these tests show that the unemployment rates in Canada and the Canadian provinces, except Nova Scotia and Quebec, do not have a unit root and are therefore, not persistent.;Essay One: Unemployment and Criminality: Using Markov-Switching Models to Identify a Link. This paper studies the existence of a relationship between the unemployment rate and crime in the U.S. Using Markov Switching Autoregressive models the behaviour of four crime variables and unemployment rate during the period of study is investigated and different regimes for each variable are determined. Using some non-parametric measures such as the Concordance Index (Harding and Pagan 2002) and Independence of Chronologies (Bodman and Crosby 2005), among others, the independency of cycles of unemployment rate and crime variables is tested. The results of this phase show that there is no relationship between the unemployment rate and burglary and motor. However, for larceny and robbery the results are mixed. At the second phase, Markov Switching Vector Autoregressive models are also used to determine the states for both unemployment rate and each of the crime variable simultaneously. The results of this stage show that the effect of unemployment rate on larceny and motor depends on the state of the variables. For larceny this effect is either positive or null, and for motor it fluctuates between negative, null, and positive. Also the result shows that regardless of the state of the variables, the effect of unemployment on burglary and robbery is negative and null, respectively.;Next, we check for stability of the persistence property of the data, using the Bai and Perron approach (1998, 2003), which allows for multiple breaks in data. The results from this section show that the persistence is not constant and is regime dependent. In addition, except for the unemployment rate in British Columbia, which is persistent in all regimes, the other series are not persistent in most regimes.