The “inner composition alignment approach” (IOTA) has been suggested in the past for detecting interdependencies between short time series. QUEST members have now published a new study with a modification of this approach (called mIOTA). The new extension overcomes the drawbacks that IOTA is unable to distinguish between positive and negative correlations, and that the null distribution for IOTA is biased towards higher values. Although the new method cannot detect the direction of the interdependencies (unlike IOTA), it outperforms standard tools for detecting interdependencies (Pearson correlation, Spearman correlation, Kendall’s τ). The method is used to derive econo-climatic networks of interdependencies between economic indicators and climatic variability for Sub-Saharan Africa and South Asia including India.
Further reading:
http://www.ias.ac.in/describe/article/conf/001/01/0051-0060