I'm working on a problem and need some help. Hope this is the right place. I ran a regression analysis on some data and came up with a model that predicts usage of 10.7209. I also computed prediction intervals for individual y. This gave me a 95% confidence level with a lower bound 9.0416 and an upper bound 12.4274. Now, what I'm having difficulties figuring out is how to construct a 95% prediction interval and use the half-width of this interval to determine whether the usage is being predicted within 10%. I'm pretty sure I take (12.4274-9.0416)/2, which gives me 1.7064. Now, I don't know what to do with this number and how to use this number with my model prediction. Can anyone give me some help? I would really appreciate it. Thanks.

Hi,

95% prediction interval is used to predict that 95% of the times next measurement will be with in the predicted interval which different from a 95% confidence interval which will be used to predict on an average that any measurement will lie with in the interval.

Regards,
Re.Vinaitheerthan