# maths - regression analysis & line of best fit?

im doing a maths thing and im a bit lost - no idea how to use the graphics calculator (T1-84 plus). i need to draw a line of best fit on a hand drawn graph and then peform a regression analysis to find the equation of the line. iv got my hand drawn graph, and now im stuck. it says to use the graphics calculator for the next part, so does anyone know what your meant to even press?

help would be awesome =)

Relevance

http://img120.imageshack.us/img120/6382/screenshot...

Notice that errors are squared so you should take squareroot if you would like to use them in a graph.

Here is matlab implemention, it might give you more insight:

function [A,B,D] = linreg(x, y)

D = det([sum(x.^2), sum(x);sum(x), length(x)]);

A = 1/D*det([sum(x.*y) sum(x); sum(y) length(x)]);

B = 1/D*det([sum(x.^2) sum(x.*y);sum(x) sum(y)]);

return

and for errors:

function [m2a m2b] = linregerror(x,y)

A = linreg(x,y);

n = length(x);

m2a = 1/(n-2)*((n*sum(y.^2)-(sum(y))^2)/(n*sum(x.^2)-(sum(x))^2)-A^2);

m2b = (sum(x))^2/n*m2a;

Edit:

Sorry didn't read your question quite well. There should be same kind of table where you can enter your datapoints. After that there will be some where option to choose line fitting. You should choose linear regression or first power line fitting.

• Login to reply the answers
• Anonymous
• Login to reply the answers