Statistische Analyse klimatologischer Zeitreihen

• Published in: Historical social research (Köln) Vol. 21; no. 3 (79); pp. 4 - 33
• Main Author: Schableger, Karl
• Format: Journal Article
• Language: German
• Published: Center for Historical Social Research / Zentrum für Historische Sozialforschung 01-01-1996
• ISSN: 0172-6404

In the scope of discussions about global warming, various statistical models describing longterm temperature time series are proposed. The time series of the daily air temperature of Vienna between 1.1.1874 and 31.12.1993 has been the basis for this analysis. In general, the results obtained from this analysis can be used for examining the climatic change of the central European area. ARIMA-Modelling, created by Box and Jenkins, was the main type of model utilized for the analysis. In addition to difference filter dynamic regression analysis and transfer function models were used. Using nonstochastic, exogenous input variables (called interventions) an expected temperature equilibrium was estimated. Moreover, ordinary trend estimation was done to get an impression of the development and power of temperature increases. Finally, the results of the analysis of extreme temperature values were suitable to the theory of global warming. Unfortunately, checking the variation of the temperature series with various measuring techniques was not an unambiguous success. Therefore, the frequently mentioned hypothesis of growing variability could not be confirmed.