Using survey-based measures of future interest rate expectations from the Blue Chip Economic Indicators and the Survey of Professional Forecasters, we examine the relationship between monetary policy uncertainty, captured as the dispersion of interest rate forecasts, and fluctuations in real economic activity and core inflation. We use a flexible time-varying parameter vector autoregression (TVP-VAR) model to clearly isolate the dynamic effects of shocks to monetary policy uncertainty. To further study possible a possible nonlinear relationship between monetary policy uncertainty and the macroeconomic aggregates, we extract the impulse-response functions (IRF’s) estimated at each quarter in the time series, and use a multi-variate regression with various measures of the shape of the IRF’s and the level of monetary policy uncertainty at that quarter in the TVP-VAR model to gauge the relationship between the effectiveness of traditional monetary policy (shocks to the Federal Funds rate), forward guidance (shocks to expected interest rates) and uncertainty. The results show that monetary policy uncertainty can have a quantitatively significant impact on output, with a one standard deviation shock to uncertainty associated with a 0.6% rise in unemployment. The indirect effects are more substantial, with a one standard deviation increase in monetary policy uncertainty associated with a 23% decrease in the maximum response of unemployment to a forward guidance episode (interest rate expectations shock). This evidence points to the importance of managing monetary policy uncertainty (clear and direct forward guidance) as a key policy tool in both stimulating economic activity as well as propagating other monetary policy through the macroeconomy.

The main focus of examination of this thesis can be broadly defined as the analysis of the main determinants of economic activity in Mexico. In this analysis, it is found that real and nominal exchange rates have enormous importance in the determination of economic activity in Mexico compared to other candidates. This conclusion is reached through a series of quantitative examinations of Mexican times series of aggregated macro economic variables. First, the determination of long-run real exchange rate is analyzed. Then, an inverse relationship between real exchange rate (defined as Pesos/Dollars) and real output and consumption has been established both empirically and theoretically, using explicit long-run models. Variance decomposition and impulse response analyses, carried out with the help of vector error correction models embedding long-run relations, have revealed the fact that the most important source of fluctuations either in consumption or output is the real exchange rate. The other variables considered in the analysis, such as supply, demand, or monetary shocks have been found to have less or non-robust importance. Next, the thesis examines the business cycle associated with exchange rate based programmes for the case of Mexico. The impulse responses, provided for this analysis, partly confirmed initial-boom-Iater-recession hypothesis observed in exchange-rate-basedstabilization programmes. The variance decomposition analysis, on the other hand, indicates that, the movements either in consumption or output can be largely explained by nominal exchange rate shocks rather than monetary shocks. Finally, given the importance of exchange rate variables, the thesis returns to the question of their determinants. Based on the quantitative analysis performed, the thesis concludes that real exchange rates can be solely explained by real shocks and nominal exchange rates can largely be explained by real shocks. Hence, it is the real exchange rate models that explain real exchange rate movements via the predominance of real shocks that get credit.

No description available.

No description available.

No description available.

This study investigates factors influencing matriculation decisions for freshman applicants in the College of Agriculture and Life Sciences (CALS) at the University of Arizona. Two different modeling approaches are used on a five-year cross-sectional sample of applicants. Consistent with previous literature, a parametric logistic regression is specified to estimate the probability that a freshman applicant will matriculate in CALS. Additionally, this study also uses non-parametric gradient boosting methods to predict whether an applicant will matriculate. As a byproduct of using two different techniques to model matriculation decisions, an additional academic interest is to see how these two distinct approaches compare in terms of explanation and predictive capabilities. The results show that students who apply early and applicants with high standardized test scores are significantly less likely to matriculate. Moreover, applicants who attend campus tours, honor students, and students from high schools with many applicants are more likely to matriculate.

This thesis investigates the relationship between discrete event and system dynamics simulation modelling. Each modelling technique has its own strengths and limitations. The choice of using one particular modelling technique often depends on the preference and the knowledge of the modeller rather than on the nature of the problem. The basis of this research has been to address the problem from a different perspective. This has been to look at the nature of the problem first, and then determine the most appropriate modelling technique to apply. The basic method adopted was to compare, contrast and experiment with these two modelling techniques in order to determine a number of common and unrelated concepts between them. This investigation discovered that a system dynamics flow diagram could be used to represent an activity cycle diagram of a discrete event model. The converted flow diagram can provide a different viewpoint from the discrete event model due to the feedback characteristic of system dynamics. This research went on to develop a set of guidelines to convert an activity cycle diagram into a system dynamics flow diagram. Experimentation with many examples demonstrates that these conversion guidelines provide a consistent and systematic method for obtaining a system dynamics flow diagram. The final stage of this research was to develop a prototype computer system (SMCP) to demonstrate these guidelines. SMCP consists of two modules, the ACD module which allows the user to specify a discrete event model through textual descriptions, and the SD module which allows the user to build a system dynamics model by drawing symbols and using text inputs. In addition, SMCP allows users to convert an ACD to a system dynamics flow diagram quickly and easy, and also demonstrates the concept of data sharing.

We are interested in the stochastic properties, individual and joint, of mis- specification testing when the data are generated by an autoregressive process. Good mis-specification tests are invariant to the dynamic properties of the pro- cess summarized by its characteristic roots, and to irrelevant misspecifications. Invariance in parameter space obviates inference prior to mis-specification test- ing. This is important as the latter is used to validate the former. Mutual independence of the tests allows calibration of the overall significance level. Es- tablishing such results requires work on individual tests and on their stochastic interactions. In Chapter 2, we derive the asymptotic distribution of two types of CUSUM of squares test, one implemented with standardized one-step-ahead OLS pre- diction errors and another implemented with OLS residuals. The latter is found to be valid in all but singular explosive cases, but the former only in purely non-explosive, or regular explosive cases with all roots in the explosive region of the parameter space; in Chapter 3, we show that a nuisance term arises in the mixed case. In Chapter 4, we derive numerically a finite sam- ple correction to render the tests implementable into software, and Chapter 1 contains two examples of applications. In Chapter 5, we consider inference on the parameters associated with the stationary part of the process, together with tests for a unit root, lag length, variance constancy, and normality of the regression innovations. In character- izing the joint distribution of these tests, we rely on asymptotic theory, and show independence in the limit. A simulation experiment suggests that finite sample correlations between some of the tests are statistically significant but small. Asymptotically, then, control of the overall significance level of the test procedure is feasible, and there is no reason to discount inference for the use of these mis-specification tests in model selection.

Recently, computational methods and software have been receiving more attention in the econometrics literature, emphasizing that they are integral components of modern econometric research. This has also promoted the development of many new econometrics software packages written in R and made available on the Comprehensive R Archive Network. This special volume on "Econometrics in R" features a selection of these recent activities that includes packages for econometric analysis of cross-section, time series and panel data. This introduction to the special volume highlights the contents of the contributions and embeds them into a brief overview of other past, present, and future projects for econometrics in R.

No description available.