Abstract
This paper presents a computational program named BINCOR (BINned CORrelation) for estimating the correlation between two unevenly spaced time series. This program is also applicable to the situation of two evenly spaced time series not on the same time grid. BINCOR is based on a novel estimation approach proposed by Mudelsee (2010) for estimating the correlation between two climate time series with different timescales. The idea is that autocorrelation (e.g. an AR1 process) means that memory enables values obtained on different time points to be correlated. Binned correlation is performed by resampling the time series under study into time bins on a regular grid, assigning the mean values of the variable under scrutiny within those bins. We present two examples of our BINCOR package with real data: instrumental and paleoclimatic time series. In both applications BINCOR works properly in detecting well-established relationships between the climate records compared.
Highlights
There are several approaches for quantifying the potential association between two evenly spaced climate time series, e.g. Pearson’s and Spearman’s correlation or the cross-correlation function (CCF)
We present a computational package named BINCOR (BINned CORrelation) that can be used to estimate the correlation between two unevenly spaced climate time series which are not necessarily sampled at identical points in time, and between two evenly spaced time series which are not on the same time grid
BINCOR is based on a novel estimation approach proposed by Mudelsee (2010)
Summary
There are several approaches for quantifying the potential association between two evenly spaced climate time series, e.g. Pearson’s and Spearman’s correlation or the cross-correlation function (CCF). The BINCOR package contains (i) a main function named bin_cor, which is used to convert the irregular time series to a binned time series; (ii) two complementary functions (cor_ts and ccf_ts) for computing the correlation between the two binned climate time series obtained with the bin_cor function; and (iii) an additional function (plot_ts) for plotting the “primary” vs the binned time series. We outline the main mathematical ideas behind the binned correlation technique for unevenly spaced sampled at different points in time, following the methodology introduced by Mudelsee (2010, 2014). The procedure is described as follows: 1. Input: two unevenly spaced climate time series {X(i), TX}iN=X1 and {Y(i), TY}iN=Y1, where TX, TY and NY, NY are the time domains and the sample sizes of each series, respectively
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call