Forecasting with serially correlated regression models

In this article we investigate the asymptotic and finite-sample properties of predictors of regression models with autocorrelated errors. We prove new theorems associated with the predictive efficiency of generalized least squares (GLS) and incorrectly structured GLS predictors. We also establish the form associated with their predictive mean squared errors as well as the magnitude of these errors relative to each other and to those generated from the ordinary least squares (OLS) predictor. A large simulation study is used to evaluate the finite-sample performance of forecasts generated from models using different corrections for the serial correlation.

Delayed traveling ionospheric disturbances, especially in equatorial regions, following geomagnetic activity

For the Western-Pacific region spread-F has been found to occur with delays after geomagnetic activity (GA) ranging from 5 to 10 days as station groups are considered from low midlatitudes to equatorial regions. The statistical (superposed-epoch) analyses also indicate that at the equator the spread-F, and therefore associated medium-scale traveling ionospheric disturbances (MS-TIDs) occur with additional delays around 16, 22 and 28 days representing a 6-day modulation of the delay period. These results are compared with similar delays, including the modulation, for D-region enhanced hydroxyl emission (Shefov, 1969). It is proposed that this similarity may be explained by MS-TIDs influencing both the F and D regions as they travel. Long delays of over 20 days are also found near the equator for airglow-measured MS-TIDs (Sobral et al., 1997). These are recorded infrequently and have equatorward motions, while normally eastward motions are measured at the equator. Also in midlatitudes D-region absorption events have been shown (statistically) to have similar long delays after GA. It is suggested that atmospheric gravity waves and associated MS-TIDs may be generated by some of the precipitations responsible for the absorption. The recording of the delayed spread-F events depends on the GA being well below the average levels around sunset on the nights of recording. This implies that lower upper-atmosphere neutral particle densities are necessary.

The specification of vector autoregressive moving average models

In this paper we propose a new identification method based on the residual white noise autoregressive criterion (Pukkila et al. , 1990) to select the order of VARMA structures. Results from extensive simulation experiments based on different model structures with varying number of observations and number of component series are used to demonstrate the performance of this new procedure. We also use economic and business data to compare the model structures selected by this order selection method with those identified in other published studies.

The relationship of cervical joint position error to balance and eye movement disturbances in persistent whiplash

Cervical joint position error (JPE) has been used as a measure of cervical afferent input to detect disturbances in sensori-motor control as a possible contributor to a neck pain syndrome. This study aimed to investigate the relationship between cervical JPE, balance and eye movement control. It was of particular interest whether assessment of cervical ME alone was sufficient to signal the presence of disturbances in the two other tests. One hundred subjects with persistent whiplash-associated disorders (WADs) and 40 healthy controls subjects were assessed on measures of cervical JPE, standing balance and the smooth pursuit neck torsion test (SPNT). The results indicated that over all subjects, significant but weak-to-moderate correlations existed between all comfortable stance balance tests and both the SPNT and rotation cervical ME tests. A weak correlation was found between the SPNT and right rotation cervical JPE. An abnormal rotation cervical JPE score had a high positive prediction value (88%) but low sensitivity (60%) and specificity (54%) to determine abnormality in balance and or SPNT test. The results suggest that in patients with persistent WAD, it is not sufficient to measure ME alone. All three measures are required to identify disturbances in the postural control system. (C) 2005 Elsevier Ltd. All rights reserved.

The roles of exercise-induced immune system disturbances in the pathology of heat stroke - The dual pathway model of heat stroke

Heat stroke is a life-threatening condition that can be fatal if not appropriately managed. Although heat stroke has been recognised as a medical condition for centuries, a universally accepted definition of heat stroke is lacking and the pathology of heat stroke is not fully understood. Information derived from autopsy reports and the clinical presentation of patients with heat stroke indicates that hyperthermia, septicaemia, central nervous system impairment and cardiovascular failure play important roles in the pathology of heat stroke. The current models of heat stroke advocate that heat stroke is triggered by hyperthermia but is driven by endotoxaemia. Endotoxaemia triggers the systemic inflammatory response, which can lead to systemic coagulation and haemorrhage, necrosis, cell death and multi-organ failure. However, the current heat stroke models cannot fully explain the discrepancies in high core temperature (Tc) as a trigger of heat stroke within and between individuals. Research on the concept of critical Tc: as a limitation to endurance exercise implies that a high Tc may function as a signal to trigger the protective mechanisms against heat stroke. Athletes undergoing a period of intense training are subjected to a variety of immune and gastrointestinal (GI) disturbances. The immune disturbances include the suppression of immune cells and their functions, suppression of cell-mediated immunity, translocation of lipopolysaccharide (LPS), suppression of anti-LPS antibodies, increased macrophage activity due to muscle tissue damage, and increased concentration of circulating inflammatory and pyrogenic cytokines. Common symptoms of exercise-induced GI disturbances include diarrhoea, vomiting, gastrointestinal bleeding, and cramps, which may increase gut-related LPS translocation. This article discusses the current evidence that supports the argument that these exercise-induced immune and GI disturbances may contribute to the development of endotoxaemia and heat stroke. When endotoxaemia can be tolerated or prevented, continuing exercise and heat exposure will elevate Tc to a higher level (> 42 degrees C), where heat stroke may occur through the direct thermal effects of heat on organ tissues and cells. We also discuss the evidence suggesting that heat stroke may occur through endotoxaemia (heat sepsis), the primary pathway of heat stroke, or hyperthermia, the secondary pathway of heat stroke. The existence of these two pathways of heat stroke and the contribution of exercise-induced immune and GI disturbances in the primary pathway of heat stroke are illustrated in the dual pathway model of heat stroke. This model of heat stroke suggests that prolonged intense exercise suppresses anti-LPS mechanisms, and promotes inflammatory and pyrogenic activities in the pathway of heat stroke.