Step-wise regression predicts cyclone movement in the southwest Pacific, suggesting improved forecasting accuracy.
Step-wise screening regression procedures have been applied to observational data on the paths of tropical cyclones in the southwest Pacific region between 5°5 and 30°5 and between 145°E and 150°W. The previous 12-hour displacement was found to be the most important predictor for future cyclone movement in both meridional and zonal directions. Terms depending on the cube of the previous 12-hour latitude or longitude displacement also appear to be statistically significant. Testing of the derived prediction equations on independent data yielded mean vector errors of approximately 109, 248 and 409 kilometres for the 12, 24 and 36-hour forecasts of cyclone displacement. These errors are of the same order as those obtained by the analog scheme of Brand and Blelloch (1976) but are rather less than the errors reported by Annette (1976) for current predictions in the Australian region.
No takes yet. Share an insight, caveat, or question.
Chong et al. (1980) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: