A hybrid linear–nonlinear approach to predict the monthly rainfall over the Urmia Lake watershed using wavelet-SARIMAX-LSSVM conjugated model

Abstract

The present study aimed to develop a hybrid model to predict the rainfall time series of Urmia Lake watershed. For this purpose, a model based on discrete wavelet transform, ARIMAX and least squares support vector machine (LSSVM) (W-S-LSSVM) was developed. The proposed model was designed to handle linear, nonlinear and seasonality of rainfall time series. In the proposed model, time series were decomposed into sub-series (approximation (a) and details (d)). Next, the sub-series were predicted separately. In the proposed model, sub-series were fed into SARIMAX to be predicted. The residual of predicted sub-series (error) of the rainfall time series was then fed into LSSVM to predict the residual components. Then, all predicted values were aggregated to rebuild the predicted time series. In order to compare results, first a classic modeling was performed by LSSVM. Later, wavelet-based LSSVM was used to capture the peak values of rainfall. Results revealed that Daubechies 4 and decomposition level 4 (db(4,4)) led to the best outcome. Due to the performance of db(4,4), it was selected to be applied in the proposed model. Based on results, it was observed that the W-S-LSSVM's performance was improved in comparison with other models.

Shipping index fitting and influencing factors extraction based on the Window-ARIMA-SVR hybrid model

“…The existing research on hybrid models can be divided into two directions 13 . First, the original time series is decomposed, each part is fitted individually using different learning models, and then the results are integrated to obtain the final one 14 . Second, the linear model has been applied to mine the linear pattern.…”

Long-term forecasting of tropical cyclones over Bay of Bengal using linear and non-linear statistical models

“…The performance of the SARIMAX model is considered better than that of the classical ARIMA model (Vagropoulos et al, 2016), though, in the past, natural disasters like earthquakes and hydroclimatic events have also been forecasted by the latter (Amei et al, 2012;Islam et al, 2021). The existing studies have highlighted the effectiveness of SARIMAX in fields like electricity, rainfall, traffic, and sales, to name a few (Elamin & Fukushige, 2018;Farajzadeh & Alizadeh, 2018). However, for forecasting cyclones, so far, its application seems to have been limited.…”

Análisis de series temporales en estaciones meteorológicas para la predicción de la precipitación en la ciudad de Manizales, Colombia

“…Sumado a esto, se han llevado a cabo diferentes estudios sobre la evolución espacial y temporal de la precipitación (Lu et al, 2019). Como ejemplo, algunos de los modelos más importantes son el modelo ARIMA que combina redes neuronales para predecir la lluvia por mes, el modelo SA-RIMA que permite el análisis de lluvia utilizando la estacionalidad en la serie de tiempo y el test de Dickey Fuller para determinar la estacionariedad (Jibril et al, 2017) y finalmente el modelo SARIMAX para predecir subseries utilizando el método de wavelet para obtener información sobre el tiempo y la frecuencia de la señal (Farajzadeh & Alizadeh, 2018). Este trabajo contribuye al pronóstico de la precipitación en la ciudad de Manizales, donde los resultados se utilizan para construir un sistema de detección temprana de deslizamientos de tierra.…”