The multispatial convergent cross mapping algorithm can be used as a test for causal associations between pairs of processes represented by time series. This is a combination of convergent cross mapping (CCM), described in Sugihara et al., 2012, Science, 338, 496-500, and dew-drop regression, described in Hsieh et al., 2008, American Naturalist, 171, 71–80. The algorithm allows CCM to be implemented on data that are not from a single long time series. Instead, data can come from many short time series, which are stitched together using bootstrapping.
|Date of publication||2014-10-19 01:55:59|
|Maintainer||Adam Clark <firstname.lastname@example.org>|
|License||GPL (>= 2)|
CCM_boot: Run multispatial CCM algorithm on two time series
ccmtest: Test for significant causal signal
make_ccm_data: Makes fake data for other functions
multispatialCCM-package: multispatial convergent cross mapping analysis
SSR_check_signal: Test process for auto-predictability.
SSR_pred_boot: State space reconstruction function