Monthly catch forecasting of anchovy Engraulis ringens in the north area of Chile:: Non-linear univariate approach

被引:62
作者
Gutierrez-Estrada, Juan Carlos
Silva, Claudio
Yanez, Eleuterio
Rodriguez, Nibaldo
Pulido-Calvo, Inmaculada
机构
[1] Univ Huelva, EPS, Dpto Ciencias Agroforestales, Palos De La Frontera 21819, Huelva, Spain
[2] Pontificia Univ Catolica Valparaiso, Fac Recursos Nat, Escuela Ciencias Mar, Valparaiso, Chile
[3] Pontificia Univ Catolica Valparaiso, Fac Ingn, Escuela Informat, Valparaiso, Chile
关键词
computational neural network; recurrent neural network; Elman model; ARIMA model; hybrid model; catch prediction;
D O I
10.1016/j.fishres.2007.06.004
中图分类号
S9 [水产、渔业];
学科分类号
0908 ;
摘要
In this study the performance of computational neural networks (CNNs) models to forecast 1-month ahead monthly anchovy catches in the north area of Chile considering only anchovy catches in previous months as inputs to the models was analysed. For that purpose several CNN approaches were implemented and compared: (a) typical autoregressive univariate CNN models; (b) a convolution process of the input variables to the CNN model; (c) recurrent neural networks (Elman model); (d) a hybrid methodology combining CNN and ARIMA models. The results obtained in two different external validation phases showed that CNN having inputs of anchovy catches of the 6 previous months hybridised with ARIMA(2,0,0) provided very accurate estimates of the monthly anchovy catches. For this model, the explained variance in the external validation fluctuated between 84% and 87%, the standard error of prediction (SEP, %) was lower than 31% and mean absolute error (MAE) was around 18,000tonnes. Also, significant results were obtained with recurrent neural networks and seasonal hybrid CNN+ ARIMA models. The strong correlation among estimated and observed anchovy catches in the external validation phases suggests that calibrated models captured the general trend of the historical data and therefore these models can be used to carry out an accuracy forecast in the context of a short-medium term time period. (C) 2007 Elsevier B.V. All rights reserved.
引用
收藏
页码:188 / 200
页数:13
相关论文
共 67 条
[1]  
Abrahart RJ, 2000, HYDROL PROCESS, V14, P2157, DOI [10.1002/1099-1085(20000815/30)14:11/12<2157::AID-HYP57>3.0.CO
[2]  
2-S, 10.1002/1099-1085(20000815/30)14:11/12&lt
[3]  
2157::AID-HYP57&gt
[4]  
3.0.CO
[5]  
2-S]
[6]   Regime shifts in the Humboldt Current ecosystem [J].
Alheit, J ;
Niquen, M .
PROGRESS IN OCEANOGRAPHY, 2004, 60 (2-4) :201-222
[7]   Evaluation of neural network streamflow forecasting on 47 watersheds [J].
Anctil, F ;
Rat, A .
JOURNAL OF HYDROLOGIC ENGINEERING, 2005, 10 (01) :85-88
[8]  
BAKUN A, 1974, FISH B-NOAA, V72, P843
[9]   Application of artificial neural networks (ANN) to primary production time-series data [J].
Belgrano, A ;
Malmgren, BA ;
Lindahl, O .
JOURNAL OF PLANKTON RESEARCH, 2001, 23 (06) :651-658
[10]  
Box G. E., 1976, TIME SERIES ANAL FOR