what is a regression line?
the best line of best fit
what is the least squares regression line?
minimising the (square) of the distances of points (residuals) from regression line
what are these regression lines used for?
1. y on x
2. x on y
- y on x used for finding the y value given an x value
- x on y is used for finding the x value given the y value
what is the formula for the least squares regression line?
* y on x
y = a+bx
what is the formula for the least squares rgression line
* x on y
x= a+ by
in the case of a random and non random variable which least squares regression line formula would we use?
- y on x
in the case of a random on random variable which least squares regression line fomrula would we use?
- both y on x and x on y
- 2 possible regression lines
what is extrapolation and interpolation?
- interpolation: within range of given x values (more reliable)
- extrapolation: outside range of given data points (less reliable coz data may change shape )
what are residuals?
on scatter graphs, the vertical distance of the point from th regression line
what is the formula for residuals?
what does the sum of square of the residuals indicate?
- measures how close the points rae to the regrssion line
- regression lines aims to minimise the sum of the square of the residuals
what is the coefficent determination?
- square of PMCC
- determines how well the regression line is a goodness of fit
- percentage of how much x determines y
Example of a normal Q
Must show all working out especially Sxx and Syy
