Steps to caluculate least value of Residual/Least error squared function/Cost function(25 September)

We will be given data points let’s say Diabetes vs Obesity.

So as per the CDC data, we have the percentage of people suffering from diabetes for each country which is differentiated by FIPS number.

What we can do is to check whether obesity is affecting diabetes or not.

So obesity will be the independent variable and diabetes will be the dependent variable.

In order to plot linear regression for diabetes against obesity.

Let’s see if we have both the points for common areas

X(Obesity)={Xp1,Xp2,Xp3……..}

Y(Diabetes)={Yp1,Yp2,Yp3,……..}

Now we must find a line that best represents the above values

line equation-.

Yi=mXi+c

where c is the y-intercept. Let’s say we draw a line from the origin so c=0

Yi=mXi

Now we have to select different slopes {m1,m2,m3……mi} or select different angles from where the line is to be drawn in the first quadrant

After that, we have to calculate the mean difference of points from the line we draw that is {Y1, Y2, Y3, Y4} to points that we obtain from the data set here it is Obesity percentage {Yp1, Yp2, Yp3, Yp4…..}

This mean difference is called a Cost function/Least Error Squared Function/Residual Error

The formula of the cost function is given by

J(mi)=1/2n*(sum of (Ypi-Yi)^2 till i reaches n)

where n is the number of points

Leave a Reply

Your email address will not be published. Required fields are marked *