Combination of inferred probabilistic models

UQ with UQLab Community Q&A and How To
1

Hi everyone,

I am currently working on a numerical model that involves four inputs, namely x1, x2, x3, and x4. My first task is to create a probabilistic model for the inputs using the inference conducted with the observed data. I would like to point out that x1 and x2 are dependent, as are x3 and x4, and there is no correlation between the two subsets of inputs.

To achieve this, I have utilized UQLab to generate two uq_input objects. One is based on the inference conducted on 100 observations of {x1, x2}, and the other on 200 observations of {x3, x4}.

I was wondering if there is a way to combine these two uq_input objects to generate a single probabilistic model of all four inputs.

Thank you very much in advance!

Best,

Hongzhou

2

Hi @HZhang,

Thank you for asking here. I think the function uq_mergeInputs does exactly what you want: in your case, you can simply call MergedInput=uq_mergeInputs(InputX1X2,InputX3X4) to combine the two input objects InputX1X2 and InputX3X4. For more advanced usage, please check the function directly or refer to uq_default_input_test_mergeInputs.m.

I hope this helps.

Best,
Xujia

1 Like
3

Hi @xujia,

This function works perfectly for my case, thank you so much!

Best,

Hongzhou