Environmetrics (2008). 19(8), 818-835
Luciana Magnano1, John W. Boland1 and Rob J. Hyndman2
- Centre for Industrial and Applied Mathematics, University of South Australia, SA 5095, Australia.
- Department of Econometrics and Business Statistics, Monash University, Clayton VIC 3800, Australia.
Abstract: We present tools to generate synthetic sequences of half-hourly temperatures with similar statistical characteristics to observed historical data. Temperatures are generated using a combination of daily and half-hourly temperature models which account for intra-day and intra-year seasonality, as well as short- and long-term serial correlations. Details of the model estimation are given as well as a description of the synthetic generation.
Keywords: temperature data, time series, Fourier series, ARMA models, seasonal block-bootstrap, synthetic generation.