We propose and evaluate explicit tests of the null hypothesis of no difference in the accuracy of two competing forecasts. In contrast to previously developed tests, a wide variety of accuracy measures can be used (in particular, the loss function need not be quadratic, and need not even be symmetric), and forecast errors can be non-Gaussian, nonzero mean, serially correlated, and contemporaneously correlated. Asymptotic and exact finite sample tests are proposed, evaluated, and illustrated.

We propose and evaluate explicit tests of the null hypothesis of no difference in the accuracy of two competing forecasts. In contrast to previously developed tests, a wide variety of accuracy measures can be used (in particular, the loss function need not be quadratic, and need not even be symmetric), and forecast errors can be non-Gaussian, nonzero mean, serially correlated, and contemporaneously correlated. Asymptotic and exact finite sample tests are proposed, evaluated, and illustrated.

We propose and evaluate explicit tests of the null hypothesis of no difference in the accuracy of two competing forecasts. In contrast to previously developed tests, a wide variety of accuracy measures can be used (in particular, the loss function need not be quadratic, and need not even be symmetric), and forecast errors can be non-Gaussian, nonzero mean, serially correlated, and contemporaneously correlated. Asymptotic and exact finite sample tests are proposed, evaluated, and illustrated.

Maximum likelihood factor analysis of time series is possible even when some series are quarterly and others are monthly. Treating quarterly series as monthly series with missing observations and replacing them with artificial observations independent of the model parameters, one can apply the Kalman filter to a state-space representation of a factor model and evaluate the likelihood function. An application to quarterly real GDP and monthly coincident business cycle indicators gives a new coincident index of business cycles. The new index is essentially the smoothed estimate of latent monthly real GDP and should improve upon the Stock-Watson index.