On the Statistical Estimation of Asymmetrical Relationship Between Two Climate Variables

Two simple methods commonly used to detect asymmetry in climate research, composite analysis, and asymmetric linear regression, are discussed compared using mathematical derivation synthetic data. Asymmetric regression is shown provide unbiased estimates only when the respective mean of positive negative events removed from both independent dependent variables (i.e., non-zero y-intercepts). Composite analysis always provides biased results strongly underestimates asymmetry, albeit less so for very larger thresholds, which cannot be with limited observational Hence, should used, even though uncertainties can large small samples. Differences estimated illustrated sea surface temperature winter level pressure signals associated El Niño La Niña.