A quadratic regression equation is of the form
y = c₃x² + c₂x +c₁
Coefficients c₁,c₂,c₃ are determined by minimizing the least squared error
![S = \sum _(i=1)^(n) [y_(i) - (c_(3)x_(i)^2+c_(2)x_(i)+c_(1))]^(2)](https://img.qammunity.org/2018/formulas/mathematics/college/n3925djmaepfcw8k1ye924my2dmt6wgxdz.png)
This means that

This creates the matrix equation for determining the c-coefficients:

For the given problem,
the matrix equation is

The solution for the coefficients is
c₁ = -0.3571
c₂ = 81.0714
c₃ = -16.4286
The graph of the fitted data is shown below.
When x = 0.25,
y = c₃(0.25)² + c₂(0.25) + c₁ = 18.884
The regression equation is
f(x) = -16.4286x² + 81.0714x - 0.3571
When rounded to the nearest whole number,
f(x) = -16x² + 81x + 0
and
f(0.25) = 19.25
Answers:
f(x) = -16x² + 81x + 0 (approximated)
f(0.25) = 19.25 (approximated)