autocorrelations

We have shown that the method used to compute certain statistics, such as autocorrelations, can lead to seriously biased results.

Trenberth [1984] « Climate Audit 2005

Nearly significant partial autocorrelations appear at other lags.

Revkin on the Hansen Fiasco « Climate Audit 2007

Maybe you will discover mid-frequency 5-10 yr cycles in stratospheric ozone that account for the observed 5-10 yr partial autocorrelations in hurricane frequency?

Hurricanes 2007 « Climate Audit 2007

In figure 1, alternative sun spot cycles are multiplied by -1, and it is this modified series that the authors claim fits with climate though autocorrelations are not considered.

Alexander et al 2007 « Climate Audit 2007

On the other hand, we show that the regression residuals from the linear detrending of a random walk series would contain large positive autocorrelations in the first few lags.

Climate Insensitivity and AR(1) Models « Climate Audit 2007

Most modern random number generators have extremely low autocorrelations, with a repetition rate measured in years, sometimes centuries or more.

Bürger Review « Climate Audit 2007

‚⠠ The station errors have large autocorrelations in time, but no correlation in space, and the grid-box sampling errors have little correlation in either space or time, so these two error components are small for global and large-scale averages.

Statistics of Record-Breaking Temperatures « Climate Audit 2007

However, your question — and the whole silly RC debacle regarding Ritson/Mann and computing the ACF correctly — touches on an important topic that Steve has written about in the past: There exist rich classes of stationary stochastic processes that provide amazingly realistic models of natural variability and exhibit lag-one autocorrelations at or near 0.

AR1 on First Differences « Climate Audit 2006

There he cites the use of a 1,3,4,3,1 filter, which will be problematic if the lag-5 autocorrelations are significant.

Bender on Hurricane Counts Continued « Climate Audit 2006

If my feeling for the statistics is correct, the main difference is that you get a better RE score even if the agreement between the datasets follows from slow autocorrelations.

Verification r2 Revealed!!! « Climate Audit 2006