Regression Lines

[butene]

How do you know if a regression line fits well with a time series graph? Do you look at its r^2 or r value? When i added a regression line to my time series graph the r^2 value is 0.902 and its r value is -0.950, so would i say it is a good fit? since the relationship between both variables does appear to be strong

[sam.utute]

To determine a good fit, look at the r value. Your value of -0.95 indicates that there is a strong negative, linear relationship between the variables.

You could say that this regression line appears to fit the data reasonably well, as shown through the Pearson' correlation coefficient of -0.95.

[butene]

Cheers Sam, but what about the r^2 value? Can't i also say that the regression line is a good fit also because that 90.2% of the variation in (dependant variable) can be explained by the variation in the (independant variable)? its R2 is also close to 1 yeah

[sam.utute]

You could use that as well. But generally, we look at the r value. It's basically common sense: if you have an r value > 0.9, then your r^2 value is going to be > 0.81. It speaks for itself. If you need to fill up lines, then go ahead and talk about it, but I don't think it's necessary.