Open AccessModel Selection For Forecasting Rainfall Dataset
Author(s) -
Amri Muhaimin,
Hendri Prabowo,
Suhartono Suhartono
Publication year - 2021
Publication title -
international journal of data science, engineering, and analytics
Language(s) - English
Resource type - Journals
eISSN - 2807-1689
pISSN - 2798-9208
DOI - 10.33005/ijdasea.v1i1.2
Subject(s) - artificial neural network , autoregressive integrated moving average , computer science , time series , feedforward neural network , machine learning , artificial intelligence , support vector machine , autoregressive model , series (stratigraphy) , data mining , statistics , mathematics , paleontology , biology
The objective of this research is to obtain the best method for forecasting rainfall in the Wonorejoreservoir in Surabaya. Time series and causal approaches using statistical methods and machine learning willbe compared to forecast rainfall. Time series regression (TSR), autoregressive integrated moving average(ARIMA), linear regression (LR), and transfer function (TF) are used as a statistical method. Feedforwardneural network (FFNN) and deep feed-forward neural network (DFFNN) is used as a machine learning method.Statistical methods are used to capture linear patterns, whereas the machine learning method is used tocapture nonlinear patterns. Data about hourly rainfall in the Wonorejo reservoir is used as a case study.The data has a seasonal pattern, i.e. monthly seasonality. Based on the cross-validation and informationcriteria, the results showed that DFFNN using the time series approach has a more accurate forecast thanother methods. In general, machine learning methods have better accuracy than statistical methods.Furthermore, additional information is obtained, through this research the parameter that best to make aneural network model is known. Moreover, these results are also not in line with the results of M3 and M4competition, i.e. more complex methods do not necessarily produce better forecasts than simpler methods.