Applicability of the MACD Technical Indicator to the Predictability of the Annual Average Flow of Paraná River

Abstract

This article assesses the applicability and effectiveness of the statistical tool “Convergence and Divergence of Moving Averages” (MACD) to the predictability of the average annual flow of the Paraná River measured in the city of Corrientes-Argentina, with a time span of 100 years. This technique is widely used in the financial market for forecasting trends and reversals in commodities and stock prices. Given the versatile nature of this technique, its parameters were easily adapted to be implemented in this study. The periods assigned to the exponential moving averages (EMA), an inherent parameter of the oscillator, allude to astronomical cycles that produce effects on the climatological variable in the addressed timescale, namely, the Schwabe Solar Cycle, with a periodicity of 11.2 years, the Lunar Nodal Cycle, of 18.6 years, and the Apsidal Lunar Cycle, of 8.85 years. There is a growing scientific understanding that such astronomical cyclical parameters do influence the terrestrial climate, especially via the modulation of ocean-atmospheric oscillations of global scale, such as the Pacific Decadal Oscillation (ODP) and El Niño-Oscilação Sul (ENOS), which are known to be related to the variable in study, which, in turn, is responsive to the rainfall conditions that supply water to two of the most important rivers in the La Plata basin: Paraná and Uruguay. Specific properties of MACD obtained considerable performance in predicting trends of increase/decrease in several periods of that centennial data series of river flow. The use of MACD for other climate variables, such as rainfall, Sea Level Pressure and Sea Surface Temperature, is highly encouraged, in order to consummate it as an additional tool to improve confidence in climate prediction.

Keywords:
climate predictability; Schwabe solar cycle; lunar cycles; flow rate; econometrics