R programming: predict(), "prediction" vs "confidence"?

Question

learning R.... anyway:

In a simple regression just x and y, I entered:

predict(data1.lm, interval="prediction")

and

predict(data1.lm, interval="confidence")

I am confused on what the difference is.

Answer 1

I don't see any parameter named "interval" in the description of confint on its help page, nor in the code for either confint.default or confint.lm. I believe this may be the root of your problem.

Responding to the edit: The prediction interval is the range in which future observation can be thought most likely to occur, whereas the confidence interval is where the mean of future observation is most likely to reside. The confidence interval is generally much more narrow than the prediction interval and its "narrowness" will increase with increasing numbers of observations, whereas the prediction interval will not decrease in width. Think 'std-error-of-the-mean' (which has a 1/N term) versus 'standard-deviation' (which only has 1/sqrt(N)). In the general case, the confidence interval and the prediction intervals will be functions of the covariates rather than just individual univariate interval pairs.

Answer 2

interval is an argument of predict, not confint.

The intervals can include the uncertainty on the estimated coefficients (confidence), the variance ("noise") in the observations, or both (prediction).