Relationship Model Anomaly Harvested Rice with a Weighted Rainfall Index in Buru Maluku Using Bootstrap Aggregating MARS Methods to Predict the Forecast Error Rates Harvested Area and Rice Production

Kondo Lembang F., Loupatty G., Talakua M. W.


Seasonal climate variations is one of the main causes of the diversity of crop production in Indonesia. Long drought and drought causing crop failures and food shortages that could affect agricultural production and food security. The indicator is a decline in acreage , harvested area and production declined sharply when climate irregularities. The magnitude of the impact caused by climatic irregularities cause we need a model that connects the harvested area with indicators of climate anomalies that can do the proper planning and anticipation measures early in order to avoid the risk of crop failure. Buru as the largest rice -producing areas in the provinces of Maluku course is expected to avoid the risk of crop failure in order not to disrupt the supply of rice. Data Collection and forecast rice production annually conducted by the Central Statistics Agency (BPS). BPS forecast model but has not entered a climatic factor, while the climate affect rice production. This research used the bootstrap aggregating MARS method to model anomaly rice harvested area with a weighted rainfall index to predict the error rate forecast harvested area and rice production. From the analysis using the best models of replication bagging MARS 150 times in the first period (January-April) and 200 times in second period (May-August) and third period (September-December) obtained an error rate forecast harvested area and rice production respectively by 5.72% and 6.81%.

Keywords— Anomaly Area harvested, weighted rainfall index, MARS, Boostrap Aggregating, rice production.

Full Text:



Breiman. L, “Bagging Predictor”, Technical report 421, Statistics Department of the University of California, 1994.

Efron. B. and Tibshirani. R.J, “An Introduction to the Bootstrap”, Chapman & Hall, Inc., New York, 1993.

Eubank. R. L, “Spline Smoothing and Nonparametric Regression”, Mercel Dekker, New York, 1988.

Friedman. J. H, “Multivariate Adaptive Regression Splines”, Rev Tech Report 102, Department of Statistics Stanford University Stanford, California, 1990.

Kartini. A. Y, “Bagging MARS for Developing forecast Models of Anomalies Area Rice Harvested of Weighted Rainfall Index in Ngawi”, Thesis, Graduate Program, Sepuluh Nopember Institute of Technology, Surabaya, 2010.

Sutikno, “Statistical Downscaling GCM Outcomes and Utilization for Rice Production Forecasting”, Dissertation, Graduate Program, Bogor Agricultural University, Bogor, 2008.

StatSoft, “Multivariate Adaptive Regression Splines (MARSplines)”, StatSoft Inc, 2000.


  • There are currently no refbacks.