Hi all,

I am trying to get minitab to display confidence interval and prediction interval in the session window for a quadratic relationship. I can see the graph and confidence intervals displayed on the graph but i don't see the actual intervals in numbers. Can minitab do that? or does it display confidence and prediction intervals for only linear models ?

Thanks in advance for any help in answering this question.

Hi all,

I am trying to get minitab to display confidence interval and prediction interval in the session window for a quadratic relationship. I can see the graph and confidence intervals displayed on the graph but i don't see the actual intervals in numbers. Can minitab do that? or does it display confidence and prediction intervals for only linear models ?

Thanks in advance for any help in answering this question.

I have not been able to get Minitab to print a list, but if you hover the mouse pointer over the line or click on the line, Minitab will display a temporary popup window with the numbers that you want. Note: If you move the mouse off the line, this list will disapear.

If you have screen capture capabilities, you should be able to capture this list in a screen shot.

Hi Miner,


I tried hovering the mouse on the confidence interval and prediction interval line and even clicking on it but it doesn't display the range or anything.
Anything i am doing wrong?


Shlagha

Here is what it looks like. I simply hovered the mouse over the confidence line and took a screen shot.

Hi all, I am trying to get minitab to display confidence interval and prediction interval in the session window for a quadratic relationship. I can see the graph and confidence intervals displayed on the graph but i don't see the actual intervals in numbers. Can minitab do that? or does it display confidence and prediction intervals for only linear models ? Thanks in advance for any help in answering this question.

I'm neither an expert on Minitab or confidence intervals around lines but is it possible to transform the quadratic equation to make it linear, get confidence intervals around that, then transform back? Have no idea of whether or not this would be a valid move from a statistical standpoint.