Online Program

Return to main conference page
Friday, May 31
Practice and Applications
Science and the Environment
Fri, May 31, 5:20 PM - 6:25 PM
Grand Ballroom J

Yield Forecasting Based on Short Time Series with High Spatial Resolution Data (306227)

Trenton Franz, University of Nebraska Lincoln 
Sayli Pokal, University of Nebraska Lincoln 
*Yuzhen Zhou, University of Nebraska Lincoln 

Keywords: Forecasting; Spatial; Short time series; Clustering

Crop yield forecasting plays an important role in planning and management of fields. Yet, it becomes especially challenging to do forecasting when we only have short time series. The corn yield data in this study were collected in high spatial resolution (i.e., 10 x 10 m spatial resolution in 800 x 800 m domain) every other year from 2002 to 2016. In this paper, we propose a clustering based spatially varying auto-regressive model for forecasting yield. We compare the forecasting performance of our model with traditional time series model and a few machine learning algorithms. The results show that for short time series with high spatial resolution data, our proposed model outperforms other models.