Title:
How Does the Temperature Vary over Time?
Evidence on the Stationary and Fractal
Nature of Temperature Fluctuations
Abstract:
This paper analyses temperature time series
data obtained from 95 worldwide weather stations,
as well as reconstructed data from the last two
millennia. We propose an empirical strategy
based on three key invariance hypotheses. The
first is that the temperature process is stationary.
The second is that the distribution of the average
temperature over any specific time period does not
depend on the length of the period, apart from
a scale transformation. The third is that the
temperature process is Gaussian. These hypotheses
imply that the temperature series follow a so-called
Fractional Gaussian Noise process, which is known
to exhibit long range dependence. In order to test
our hypotheses we have applied non-parametric graphical
methods and Chi-square tests. The tests indicate
that both the observed and the reconstructed data
are consistent with the Fractional Gaussian Noise model.
John Dagsvik: How Does the Temperature Vary over Time? Evidence on the Stationary and Fractal Nature of Temperature Fluctuations
John Dagsvik ( Statistics Norway ) gives a seminar in room 107, 1st floor N.H. Abels House at 14:15 March 10th: How Does the Temperature Vary over Time? Evidence on the Stationary and Fractal Nature of Temperature Fluctuations
Published Feb. 24, 2015 3:56 PM
- Last modified Feb. 24, 2015 6:28 PM