On typical characteristics of economic time series and the relative qualities of five autocorrelation tests

Bias-adjusted estimation in the ARX(1) model

“…To further investigate the robustness of our findings, various other X-matrix specifications were tried, including (i) use of boolean (dummy) vectors, as would be used, for example, when working with models with outliers or structural breaks, and (ii) the matrix E k , specified as the first k eigenvectors of the first order difference matrix, for various values of k. Matrix E k is given by x it = cos[(2t − 1)π(i − 1)/(2T )] (see Durbin and Watson, 1971), and is a useful benchmark because these vectors tend to mimic the behavior of economic time series with seasonal and cyclical-type behavior (Dubbelman, Louter and Abrahamse, 1978;King, 1985, p. 32). For all X-matrices considered, the ranges for which the respective estimators perform best were virtually identical.…”

Autoregressive Lag—Order Selection Using Conditional Saddlepoint Approximations

“…(Durbin andWatson 1950, 1971) and the references therein), the optimality of the test breaks down in either the non-circular model and/or with arbitrary exogenous X, but does provide strong motivation for an approximately UMPU test in the general setting considered here. This is particularly so for economic time series, as they typically exhibit seasonal (i.e., cyclical) behavior that can mimic the Fourier regressors in (8) (Dubbelman et al 1978;King 1985, p. 32).…”

Nonnested Testing for Competing Autoregressive Dynamic Models Estimated by Instrumental Variables

“…Dubbelman et al (1978) and Hendry and Richardson (1983) have therefore stressed the need for an assumed model to encompass competing specifications, in other words, the need for a null model to be able to predict the performance of viable alternative specifications "significantly well." This is the main role of nonnested testing.…”

“…Their basic ideas have been adapted to linear regression models by Pesaran (1974) and to nonlinear regression models by Pesaran and Deaton (1978). Dubbelman et al (1978), McAleer (1987), and Berger et al (1988) provided excellent introductions to the theory and application of nonnested tests. Other relevant works in this area include those developed by Pesaran and Pesaran (1993), Huth et al (1993), Silvapulle and King (1993), McAleer (1995), Fan and Li (1995), Godfrey (1998), Berger and Mortera (1999), Davidson and Mackinnon (1981, 2002, Timm and Al-Subaihi (2001), and Watnik et al (2001).…”