RoXie

So, of course, this may be confused by the lines of communication, but assuming I understand the problem and got the right X and Y, I come up with a model that looks like yours:

## 
## Call:
## lm(formula = dat$Year ~ dat$MonthMean)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -11.654  -2.274   0.641   4.124   7.718 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    1980.54       2.39  828.81  < 2e-16 ***
## dat$MonthMean    10.12       1.39    7.29  6.1e-08 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 
## 
## Residual standard error: 5.26 on 28 degrees of freedom
## Multiple R-squared: 0.655,   Adjusted R-squared: 0.643 
## F-statistic: 53.2 on 1 and 28 DF,  p-value: 6.1e-08 
## 

Which could be plotted as

plot of chunk reg2

And the only thing I see different from yours is you seem to have your Y and X axes reversed, so you have a plot that reflects the true plot about the y=x line. So I would suggest Y=Year on the up and down / Y axis and X=Mean Price per Gallon on the right/left X-axis.

Additionally, I see an equation on your graph that says \[ y=.0648x -127.74 \] whereas the equation of the regression line should be \[ y=1980.54 + 10.123x \] (which you can find from the output from your regression) Perhaps this is a result of your axes switch?

Hey is that what you needed?