Optimization cannot proceed because the initial value of the negative log-likelihood is NaN. This can occur when the in

[shariari]

Optimization cannot proceed because the initial value of the negative log-likelihood is NaN. This can occur when the initial parameter estimates are extremely poor or when selecting an error model that is inappropriate for your data. Ensure that initial parameter estimates result in reasonable simulation results. When using the proportional error model, ensure that simulation results are never zero. When using the exponential error model, ensure that simulation results are always positive.