Testing and estimation of spatial econometric model in the case of limited samples

In econometric models, the error term is usually assumed to be normally distributed for ease of calculation. However, in the real socio-economic world, the distribution of error terms is usually unknown or does not satisfy the classical assumption of normal distribution. The random sampling method provides an effective way to solve the econometric model study of the unknown or non-normal distribution of error terms. The Monte Carlo method and Bootstrap method are often used to solve related problems in the case of limited samples.