Time-varying combinations of predictive densities using nonlinear filtering

We propose a Bayesian combination approach for multivariate predictive densities which relies upon a distributional state space representation of the combination weights. Several specifications of multivariate time-varying weights are introduced with a particular focus on weight dynamics driven by the past performance of the predictive densities and the use of learning mechanisms. In the proposed approach the model set can be incomplete, meaning that all models can be individually misspecified. A Sequential Monte Carlo method is proposed to approximate the filtering and predictive densities. The combination approach is assessed using statistical and utility-based performance measures for evaluating density forecasts of simulated data, US macroeconomic time series and surveys of stock market prices. Simulation results indicate that, for a set of linear autoregressive models, the combination strategy is successful in selecting, with probability close to one, the true model when the model set is complete and it is able to detect parameter instability when the model set includes the true model that has generated subsamples of data. Also, substantial uncertainty appears in the weights when predictors are similar; residual uncertainty reduces when the model set is complete; and learning reduces this uncertainty. For the macro series we find that incompleteness of the models is relatively large in the 1970’s, the beginning of the 1980’s and during the recent financial crisis, and lower during the Great Moderation; the predicted probabilities of recession accurately compare with the NBER business cycle dating; model weights have substantial uncertainty attached. With respect to returns of the S&P 500 series, we find that an investment strategy using a combination of predictions from professional forecasters and from a white noise model puts more weight on the white noise model in the beginning of the 1990’s and switches to giving more weight to the professional forecasts over time. Information on the complete predictive distribution and not just on some moments turns out to be very important, above all during turbulent times such as the recent financial crisis. More generally, the proposed distributional state space representation offers great flexibility in combining densities.     

Is social protection a necessity or a luxury good? New multivariate cointegration panel data results

The aim of this paper is to test the claim that social protection is a luxury good. Therefore, GDP elasticity of selected social protection expenditure is estimated using a new econometric approach developed first by Kao and Chiang (Advances in Econometrics, 15, 179–222, 2000). Time series properties of selected social expenditure in 18 OECD countries from 1981 to 1998 are examined. Using panel data cointegration tests and OLS, FMOLS and DOLS estimators, results were found which differ from previous analyses reporting substantially higher income elasticities. With the FMOLS, selected social expenditure has income elasticities smaller than one but greater than one with the DOLS. It is noteworthy that whether selected social expenditure is stationary or nonstationary may have critical implications for researchers and policy makers desiring to model and explain the impact of this expenditure on a country economic system.     

Do rainfall conditions push or pull rural migrants: evidence from Malawi

This article uses nationally representative data from Malawi's 2004/05 Integrated Household Survey (IHS2) to examine whether rainfall conditions influence a rural worker's decision to make a long‐term move to an urban or another rural area. Results of a Full Information Maximum Likelihood regression model reveal that (1) rainfall shocks have a negative association with rural out‐migration, (2) migrants choose to move to communities where rainfall variability and drought probability are lower, and (3) rainfall shocks have larger negative effects on the consumption of recent migrants than on the consumption of long‐time residents.     

A family portrait of Canada's most successful female entrepreneurs

In an attempt to study the factors contributing to the decision to become an entrepreneur, an intensive interview survey of 36 successful women entrepreneurs was conducted. The importance of paternal occupation and psychodynamic interactions with both the mother and father was highlighted. The study revealed mirror images of the patterns found to be correlated with male entrepreneurship.Dr. Belcourt is an Associate Professor, Atkinson College, York University. Formerly, she held positions at Concordia University and Universite du Quebec a Montreal. Prior to her university career, she was Manager, Personnel, CP Rail, and held various management positions with the Federal Government.     

Experiments in scientific computation on the PlayStation 3

The Sony PlayStation 3 (PS3) offers the computational power of a parallel processor at low cost, which makes it a great starter unit for development in parallel programming. To explore the capabilities of the unit, we took a simple ray tracing program and extended it to render triangulated height field data across the PS3’s 6 synergistic processing units (SPUs). We also implemented the heat averaging equation as a precursor to CFD analysis on the PS3. In our studies, we found the Cell engine in the PS3 to be a powerful machine, however great care must be taken while developing because its unique platform calls for many levels of optimization to ensure efficiency.     

Special interest groups and the allocation of public funds

I propose a fully rational model of government contracting that explains differences in local government spending from grants and other income. In this model, violations of fungibility arise from dynamic interactions between politicians and interest groups with the ability to raise funds for the local government. The predictions of the model are tested by exploiting unique features of windfalls received by states under a settlement with the tobacco industry. Although windfalls are legally unrestricted, the median state increased spending on tobacco control programs from zero to $2.30 per capita upon receipt of funds. The marginal propensity to spend on such programs is 0.20 from settlement revenue and zero from overall income. States which were not involved in the settlement lawsuits spend less. These results cannot be explained by existing models in the literature.     

Who’s calling? The effect of phone calls and personal interaction on tax compliance

International Tax and Public Finance - Most tax agencies use letters as the method of communicating with taxpayers. Still, other technologies exist that could be more effective. This paper reports...     

HR Practices,Customer-Focused Outcomes,and OCBO: The POS-Engagement Mediation Chain

Employee Responsibilities and Rights Journal - The mechanisms through which HR practices impact employee-level and organizational-level outcomes continue to intrigue HR academicians and...