227k views
2 votes
find the least squares straight line y = mx b to fit the data points: (1, 9),(2, 7),(3, 3),(4, 2). compute the minimum square error.

User Aladdin
by
8.2k points

1 Answer

5 votes
Answer: 67.315

We start by using the formula for the slope of the least squares regression line:

m = (nΣ(xy) - ΣxΣy) / (nΣ(x^2) - (Σx)^2)

where n is the number of data points, and Σ denotes the sum of the indicated quantities.

Substituting the given values, we get:

m = (4(1x9 + 2x7 + 3x3 + 4x2) - (1+2+3+4)(9+7+3+2)) / (4(1^2 + 2^2 + 3^2 + 4^2) - (1+2+3+4)^2)
= (-39) / 20
= -1.95

Next, we use the formula for the y-intercept:

b = (Σy - mΣx) / n

Substituting again, we get:

b = (9+7+3+2 - (-1.95)(1+2+3+4)) / 4
= 6.425

So the equation of the least squares regression line is:

y = -1.95x + 6.425

To compute the minimum square error, we use the formula:

SSE = Σ(y - yhat)^2

where SSE stands for sum of squared errors, y is the actual y-value of each data point, and yhat is the predicted y-value on the regression line.

Substituting the given values, we get:

SSE = (9 - (-1.95)(1) + 6.425)^2 + (7 - (-1.95)(2) + 6.425)^2 + (3 - (-1.95)(3) + 6.425)^2 + (2 - (-1.95)(4) + 6.425)^2
= 67.315

Therefore, the minimum square error is approximately 67.315.
User Houshalter
by
8.3k points

No related questions found