Use regression models to predict an independent variable

Question

I am currently developing regression models to predict a variable y, which is a function of the variables X1, X2, ..., Xn.

However, in the practical application of the models, one of the X variables needs to be calculated based on a fixed value for y. What would be the best way to achieve this result? In parametric models I assume this could be achieved mathematically, but what about non-parametric models? Does it make sense to train the model considering the variable X as the result and y as one of the input variables?

Thanks!

Answer 1

If you take y as a input feature to predict X variable it wont contribute anything for model since you have fixed value for y variable. It will end up with zero standard deviation. I would suggest take remaining X1,X2.. variables as input features and predict for X variable.