The Student Room Group

Regression Curves

I know there is an equation for regression lines, but is there a similar formula for regression curves where a simple line wouldn't approximate the trend of the data very well?

For example, if the underlying pattern in the data suggested a quadratic or cubic trend, would there be a specific equation to work out the regression curve for this set of data (assuming that regression curves do exist :tongue:)

Thanks for helping! :smile:

Posted from TSR Mobile
Reply 1
Original post by majmuh24
I know there is an equation for regression lines, but is there a similar formula for regression curves where a simple line wouldn't approximate the trend of the data very well?

For example, if the underlying pattern in the data suggested a quadratic or cubic trend, would there be a specific equation to work out the regression curve for this set of data (assuming that regression curves do exist :tongue:)

Thanks for helping! :smile:

Posted from TSR Mobile


Tempted as I am to link you to my effort I have to admit that this is far better.

You can see it in action and read about how it works. :smile:

Quick Reply

Latest