- Thread starter Cynderella
- Start date
- Tags accuracy inference nonparametric parametric data unbiased estimator

Another researcher estimated the same parameter in non-parametric way.

But I don't understand how it is possible to estimate a parameter in a non-parametric way, i.e. without assuming the existence of a parameter.

I believe that I understand that it is possible to estimate a parameter with parametric methods.

But I don't understand how it is possible to estimate a parameter in a non-parametric way, i.e. without assuming the existence of a parameter.

But I don't understand how it is possible to estimate a parameter in a non-parametric way, i.e. without assuming the existence of a parameter.

regards

Yepp, one difference might be that the parametric method will estimate a mean. The non-parametric method most probably the median . If the distribution is not symmetrical there will be a difference.

You can get the median from minimizing the absolute deviation (in contrast to least squares to get the mean). But isn't that a parametric method that estimates the parameter theta by minimizing Sum|y-theta| ?