Auto-covariance function

Climate of the Earth system

Prof. Dr. Markus Meier
Leibniz Institute for Baltic Sea Research Warnemünde (IOW)
E-Mail: markus.meier@io-warnemuende.de

Auto-covariance function#

  • auto-covariance function γ describes the time scale behavior of a time series: how likely a variable depends on its own time series. an example of autocorrelation: if todays weather is warm there is a high likelyhood that tomorrows weather is also warm

  • non-parametric estimator of the auto-correlation function:

ρ(τ)=γ(τ)γ(0),
  • where γ(τ) is the sample auto-covariance function, (covariance of X with itself at a time lag of τ)

γ(τ)=1Tt=|τ|+1TXt|τ|Xt
  • characteristics of γ:

    • γ is symmetric

    • γ(0)=Var(X)

    • γ(τ)γ(0)

    • |ρ(τ)|1

    • ρ(τ)<0  oscillation

  • ρ(τ) illustrates how the dynamical system responds to a disturbance from the equilibrium, persistence of a forecast skill

../_images/L15_1_temperature_series.PNG
../_images/L15_3_elnino.PNG
  • auto-correlation functions of the above time series in Figure 3 below

../_images/L15_2_acf.PNG

Autoregressive (AR) processes#

  • in an AR process each value in a time series is modeled as a linear combination of its previous values, along with some random noise. a general AR(p) process can be represented as follows:

Xt=α1Xt1+α2Xt2+...αpXtp+εt

α1,α2,...,αp - autoregressive coefficients determining the dependance on the past values,
εt - random error term or white noise at time t

  • order p of the autoregressive process determines how many past values have to be considered in the predicition of the current value

  • AR(1) processes rely only on the previous value

Xt=αXt1+εt,
  • while AR(2) processes rely on the previous two values

Xt=α1Xt1+α2Xt2+εt,
../_images/L15_4_AR1.png
../_images/L15_5_AR2.png

Decorrelation time#

  • describes characteristic timescales of stochastic processes: how fast the autocorrelation decreases with time lag as a measure of the memory and persistence of the processes

  • number degree of freedom nX of X (initially we had the definition for nX that Xi and Xj have to be independent, meaning uncorrelated), central limit theorem:

Var(X¯)=σX2nXnX
  • relation between the true number of time steps used and nX is the decorrelation time

τD=nnX=1+2k=1ρ(k),     for the mean=1+2k=1ρ(k)2,   for the variance