still trying in R with CI and predictions

Question

I looked over the tutorial in R for prediction and decided to reword my question. I need the prediction on average for male with a 95%CI then repeat with maximal values using the variables status, income and verbal. Now If I did the predictions correctly shown below with the mean then the max values. But doesn't that give 2 prediction intervals and 2 confident intervals for both the average and max. values

 g2<-data.frame(status=75, income=15, verbal=10, sex=0)
 g1<-data.frame(status=43, income=4.64, verbal=6.66, sex=0)

ie..Average

> g1<-data.frame(status=43, income=4.64, verbal=6.66, sex=0)
> predict(g,g1, interval='confidence', level=.90)
fit lwr upr
1 28.11506 19.7603 36.46983
> predict(g,g1, interval='prediction', level=.90)
fit lwr upr
1 28.11506 -10.95281 67.18293

Max Values used:

> predict(g,g2, interval='confidence', level=.90)
       fit      lwr      upr
1 71.30794 47.07516 95.54072
> predict(g,g2, interval='prediction', level=.90)
       fit      lwr      upr
1 71.30794 26.10037 116.5155

Do I need both the prediction and confidence interval?

Question: how do I plot the CI to determine which is wider?

Answer 1

Do I need both the prediction and confidence interval?

It depends on the question you've been asked for this homework. A confidence interval tells you about the likely location of the true population parameter (you may recall that you are fitting a mean to a sample of data), whereas a prediction interval is about the distribution of individual data points, specifically newly sampled observations.
(Sidenote: They both differ from a tolerance interval.)

In sum, are you interested in getting an interval for future observations or not? The beginning of your question speaks of itself.

Related threads (besides https://stats.stackexchange.com/q/37959/930):

• Difference between confidence intervals and prediction intervals
• Confidence and prediction intervals of linear regression model
• Obtaining a formula for prediction limits in a linear model

(which also suggest you may want to have a look at past threads on this site before asking close to duplicate question.)

Note also that you asked for a 90% CI in your code (level=.90), not a 95% CI as stated at the beginning of your question.

Question: how do I plot the CI to determine which is wider?

You don't need to plot them to verify which one is larger. Theory and intuition already tell you that: would you be more confident with what you estimated from observed data, or from unseen statistical units?

In any case, there is an example of plotting confidence interval in R on-line help, see help(predict.lm); however, as pointed out by @whuber in his comment, you must be careful with object names in R because R is case-sensitive.