US Patent 11062223 Forecasting field level crop yield during a growing season

A method for predicting field specific crop yield recommendations is disclosed. A computer system receives data records, including remotely sensed spectral property of plant records and soil moisture records. The system aggregates the records to create geo-specific time series over a specified time. The system selects representative features from the geo-specific time series and creates, for each specific geographic area, a covariate matrix in computer memory comprising the representative features. The system assigns a probability value to a component group in a set of parameter component groups, where each component group includes one or more regression coefficients and an error term calculated from probability distributions. The system is programmed to generate generates the probability distributions used to determine the regression coefficients and the error term, the probability distribution used to generate the error term is defined with a mean parameter set at zero and a variance parameter set to a field specific bias coefficient.