The problem of predicting 0-1-events is considered under general conditions, including stationary processes with short and long memory as well as processes with changing distribution patterns. Nonparametric estimates of the probability function and prediction intervals are obtained.

SEMIFAR models introduced in Beran (1999) provide a semiparametric modelling framework that enables the data analyst to separate deterministic and stochastic trends as well as short- and long-memory components in an observed time series. A correct distinction between these components, and in...

A limit theory is developed for the least squares estimator for mildly and purely explosive autoregressions under drifting sequences of parameters with autoregressive roots ρn satisfying ρn → ρ ∈ (-∞, -1] ∪ [1, ∞) and n (|ρn| -1) → ∞. Drifting sequences of innovations and...

This paper applies a local-linear non-parametric kernel regression technique to examine the effect of macroeconomic factors on stock market performance in Ghana. We show that the popular parametric specification in the existing literature suffers from functional misspecification. The evidence...

Weather and temperatures vary in ways that are difficult to explain and predict precisely. In this article we review data on temperature variations in the past as well possible reasons for these variations. Subsequently, we review key properties of global climate models and statistical analyses...