Time series analysis of meteorological data : wind speed and direction

by Pang, Wan-kai

Abstract (Summary)
(Uncorrected OCR) ABSTRACT Statistical literature about time series modelling on wind speed and direction are scarce. As far as weather forecasting is concerned, a thorough understanding of the time dependence structure of the two variables are of great importance since it could well avoid making erroneous forecasts. This thesis considers the application of some time series methods to wind speed and direction. Firstly we concentrate on analysing wind speed and a thorough treatment is gone through from distribution fitting to time series modelling and forecasting future wind speed. Subsequently, we go on to analyse wind direction data alone. This is an area which is rather new in the field of time series and forecasting. A detailed analysis on this kind of directional data is done. Again we take a different approach in dealing with those two variables and come up with some nice properties. Furthermore investigations about the lag-dependence about wind speed and direction are carried out and we introduce an alternative method to detect lag-dependence. Finally hypothesis testing of complex time series is considered in the last chapter with application to real data.
Bibliographical Information:


School:The University of Hong Kong

School Location:China - Hong Kong SAR

Source Type:Master's Thesis

Keywords:winds speed measurement time series analysis


Date of Publication:01/01/1993

© 2009 All Rights Reserved.