bootstrap sampling

[Grogu1992]

I am applying "probabilistic reconciliation" using the bootstrap intervals_method. I have been trying to replicate the method’s results for different sample sizes; however, I only get exactly the same results when the sample size is set to 200.

I reviewed the code for the reconcile method and noticed that a default value of 200 is assigned to the reconciler's arguments.

Is this correct? If so, is there any explanation for it?
Is there any way to change this parameter using the input arguments of the reconcile method?

Thanks

[elephaint]

I am applying "probabilistic reconciliation" using the bootstrap intervals_method. I have been trying to replicate the method’s results for different sample sizes; however, I only get exactly the same results when the sample size is set to 200.

Do you have code that I can run that shows the difference?

I reviewed the code for the reconcile method and noticed that a default value of 200 is assigned to the reconciler's arguments.

Is this correct? If so, is there any explanation for it? Is there any way to change this parameter using the input arguments of the reconcile method?

Thanks

It's correct; it's because it can be the case that num_samples isn't provided but level is, and then we need n_samples to be positive. However, that can be more easily achieved I think, so I think we will change this behavior.

No, atm you can't change that parameter.