Search

The author proposes a class of exact tests of the null hypothesis of exchangeable forecast errors and, hence, of the hypothesis of no difference in the unconditional accuracy of two competing forecasts.

The authors examine evidence of long- and short-run co-movement in Canadian sectoral output data. Their framework builds on a vector-error-correction representation that allows them to test for and compute full-information maximum-likelihood estimates of models with codependent cycle restrictions.

Debt strategy is defined as the manner in which a government finances an excess of government expenditures over revenues and any maturing debt issued in previous periods. The author gives a thorough qualitative description of the complexities of debt strategy analysis and then demonstrates that it is, in fact, a problem in stochastic optimal control.

The author evaluates the ability of a variety of output-gap estimators to accurately measure the output gap in a model economy. A small estimated model of the Canadian economy is used to generate artificial data.

Postulating two different specifications for the Canadian Phillips curve (a purely backwardlooking model, and a partly backward-, partly forward-looking model), the authors test for structural breaks in the parameters of the equation. In each case, they account for the possibilities that: (i) breaks can be discrete, or continuous, and (ii) available data samples may be too small to justify using asymptotically valid structural-change tests.

This paper shows how existing band-pass filtering techniques and their extension can be applied to the common current-analysis problem of estimating current trends or cycles.

The exponential family, relative entropy, and distortion are methods of transforming probability distributions. We establish a link between those methods, focusing on the relation between relative entropy and distortion.

Wavelets are mathematical expansions that transform data from the time domain into different layers of frequency levels. Compared to standard Fourier analysis, they have the advantage of being localized both in time and in the frequency domain, and enable the researcher to observe and analyze data at different scales.

This paper describes a new test for evaluating conditional density functions that remains valid when the data are time-dependent and that is therefore applicable to forecasting problems. We show that the test statistic is asymptotically distributed standard normal under the null hypothesis, and diverges to infinity when the null hypothesis is false.

This paper evaluates linear and non-linear forecast-combination methods. Among the non-linear methods, we propose a nonparametric kernel-regression weighting approach that allows maximum flexibility of the weighting parameters.