Even in scientific disciplines, forecast failures occur.  Four possible states of nature (a model is good or bad, and it forecasts well or badly) are examined using a forecast-error taxonomy, which traces the many possible sources of forecast errors.  This analysis shows that a valid model can...

Success in accurately forecasting breaks requires that they are predictable from relevant information available at the forecast origin using an appropriate model form, which can be selected and estimated before the break.  To clarify the roles of these six necessary conditions, we distinguish...

Unpredictability arises from intrinsic stochastic variation, unexpected instances of outliers, and unanticipated extrinsic shifts of distributions.  We analyze their properties, relationships, and different effects on the three arenas in the title, which suggests considering three associated...

We develop forecast-error taxonomies when there are unmodeled variables, forecast 'off-line'.  We establish three surprising results.  Even when an open system is correctly specified in-sample with zero intercepts, despite known future values of strongly exogenous variables, changes in...

Although a general unrestricted model may under-specify the data generation process, especially when breaks occur, model selection can still improve over estimating a prior specification.  Impulse-indicator saturation (IIS) can 'correct' non-constant intercepts induced by location shifts in...

The new-Keynesian Phillips curve (NKPC) includes expected future inflation as a major feedforward variable to explain current inflation.  Models of this type are regularly estimated by replacing the expected value by the actual future outcome, then using Instrumental Variables or Generalized...

Trygve Haavelmo's Probability Approach aimed to implement economic theories, but he later recognized their incompleteness. Although he did not explicitly consider model selection, we apply it when theory-relevant variables, {xt}, are retained without selection while selecting other candidate...

In a non-stationary world subject to structural breaks, where model and mechanism differ, equilibrium-correction models are a risky device from which to forecast. Equilibrium shifts entail systematic forecast failure, and indeed forecasts will tend to move in the opposite direction to the data....

We consider forecasting using a combination, when no model coincides with a non-constant data generation process (DGP). Practical experience suggests that combining forecasts adds value, and can even dominate the best individual device. We show why this can occur when forecasting models are...

Parallel computation has a long history in econometric computing, but is not at all wide spread. We believe that a major impediment is the labour cost of coding for parallel architectures. Moreover, programs for specific hardware often become obsolete quite quickly. Our approach is to take a...