First, consider whether it is possible to include in the dataset individuals with unknown location (just by not considering their contribution to the spatial model), like it is for AR.
That would be useful for prediction.
In the meanwhile, check for this and give a more explicit error message, indicating the offending line in the coordinates.
First, consider whether it is possible to include in the dataset individuals with unknown location (just by not considering their contribution to the spatial model), like it is for AR. That would be useful for prediction.
In the meanwhile, check for this and give a more explicit error message, indicating the offending line in the coordinates.