No abstract

No abstract

This paper is concerned with tests in multivariate time series models made up of random walk (with drift) and stationary components. When the stationary component is white noise, a Lagrange multiplier test of the hypothesis that the covariance matrix of the disturbances driving the multivariate random walk is null is shown to be locally best invariant, something that does not automatically follow in the multivariate case. The asymptotic distribution of the test statistic is derived for the general model. The test is then extended to deal with a serially correlated stationary component. The main contribution of the paper is to propose a test of the validity of a specified value for the rank of the covariance matrix of the disturbances driving the multivariate random walk. This rank is equal to the number of common trends, or levels, in the series. The test is very simple insofar as it does not require any models to be estimated, even if serial correlation is present. Its use with real data is illustrated in the context of a stochastic volatility model, and the relationship with tests in the cointegration literature is discussed.

The locally most powerful test is derived for the hypothesis that the regression coefficients are constant over time against the alternative that they vary according to the random walk process. When the regression equation contains the constant term only, comparisons are made with the tests suggested by LaMotte and McWhorter (1978). These are based on exact powers and on three different types of asymptotic efficiencies including the classical Pitman and Bahadur approaches and the new one due to Gregory (1980). The concept of the Bahadur efficiency is extended to cover also the random slopes. Suggestions are made for choosing the test.

General methodology is developed here to deal with the association between a a binary variable and network connections with or without confounding covariates. Also the case when the network is observed at several time periods is treated. As an application we consider the diffusion of organic farming in the province of North Karelia in Finland. It turns out that organic farms are more clustered than would be expected under pure random allocation. The neighborhood effect remains when adjusting for the production lines of the farms. The spatio-temporal analysis shows that new adopters are more often found within the neighborhoods of each others and of earlier adopters.