Premium
Data Generation Processes for Spatial Series: The Example of Ephemeral Channel Form
Author(s) -
Alexander C. O.,
Rendell H. M.
Publication year - 1995
Publication title -
geographical analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.773
H-Index - 65
eISSN - 1538-4632
pISSN - 0016-7363
DOI - 10.1111/j.1538-4632.1995.tb00337.x
Subject(s) - ephemeral key , channel (broadcasting) , series (stratigraphy) , time series , autoregressive model , data series , vector autoregression , computer science , econometrics , geology , statistics , algorithm , mathematics , telecommunications , paleontology
Recently developed techniques for time series analysis are applied to channel width and slope series collected in small ephemeral channels in Southern Italy. Tests for the difference between trend and difference stationarity, although developed in the field of econometrics, are found to be more suitable for data generation processes in geomorphology, where physical causes of deterministic trends are evident. These tests are important because the modeling of causal flows and development of dynamic predictive models depends on the type of data generation process that is identified. The ephemeral channel data exhibit trend stationarity and the results of spectral analysis and vector autoregression analysis of detrended channel width and slope data are described.