### Interval regression

Posted:

**Tue Apr 10, 2018 12:54 pm**Hello everybody,

I have a question regarding an estimation procedure and I'm not quite sure about which one to use. I guess it could be some sort of an interval or quantile regression. One of my independent variable is an estimate and can take continous values between 0 and 1. We can't say for sure if this estimate matches the real value but we can say for sure that this estimate is the absolute minimum for the real value.

For example:

X = 0.21

It is possible that 0.21 doesn't match the real value but we know for sure that the real value is within the interval of 0.21 and 1. So the estimate is the lower bound of an interval.

Is there some estimation procedure that makes it possible to use intervals as independent variables? I read something about interval regression, however in those examples only the dependent variable was an interval.

I hope someone can help me!

Cheers,

png1

I have a question regarding an estimation procedure and I'm not quite sure about which one to use. I guess it could be some sort of an interval or quantile regression. One of my independent variable is an estimate and can take continous values between 0 and 1. We can't say for sure if this estimate matches the real value but we can say for sure that this estimate is the absolute minimum for the real value.

For example:

X = 0.21

It is possible that 0.21 doesn't match the real value but we know for sure that the real value is within the interval of 0.21 and 1. So the estimate is the lower bound of an interval.

Is there some estimation procedure that makes it possible to use intervals as independent variables? I read something about interval regression, however in those examples only the dependent variable was an interval.

I hope someone can help me!

Cheers,

png1