The analysis of directional time series: Applications to wind speed and direction. (Based on the author's thesis, Univ. of Western Australia in Perth) (Q1188593)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | The analysis of directional time series: Applications to wind speed and direction. (Based on the author's thesis, Univ. of Western Australia in Perth) |
scientific article |
Statements
The analysis of directional time series: Applications to wind speed and direction. (Based on the author's thesis, Univ. of Western Australia in Perth) (English)
0 references
17 September 1992
0 references
The subject of this study is the analysis of directional time series. It has been divided into two parts. Part I contains an analysis of a special two-dimensional wind series. The dimensions are wind speed and direction. This series is regarded as superposition of several components: A geostrophic component, a land and sea breeze cycle, a storm component and a residual component of short term fluctuations. The determination of the components via robust filtering of the series is pursued after an introduction to general weather patterns has been given and some robust filter study has been presented. The theoretical background necessary for the analysis of directional time series is presented in Part II. This part focusses on the discussion of two parametric models for directional time series. The models adopt the standard autoregressive (AR-) scheme and are called von Mises and wrapped AR-processes, respectively. Most of the discussion in Part II is devoted to fitting a wrapped AR-model to a time series of angular data. Concepts of angular dependency and a circular autocovariance function are developed. Given the circular ACF of a wrapped AR-process the estimation of the underlying unwrapped process is considered. At last, the concepts are applied to the analysis of the residual component of the wind directions.
0 references
autoregressive scheme
0 references
von Mises process
0 references
directional time series
0 references
two- dimensional wind series
0 references
robust filtering
0 references
wrapped AR-processes
0 references
time series of angular data
0 references
angular dependency
0 references
circular autocovariance function
0 references