Why can linear approximation equal quadratic approximation

In summary, the reason why the linear approximation of 1/(1-u) where u = x2 is the same as the quadratic approximation of 1/(1-x2) is due to the composition laws for polynomial approximations. This means that the polynomial approximations of two functions, when composed together, result in a polynomial approximation at the same order. In this case, the linear approximation of the function 1/(1-u) is equivalent to the quadratic approximation of the function 1/(1-x2) because of the composition of the functions x^2 and 1/(1-u).
  • #1
JS-Student
10
0
Hi

I'm having trouble visualizing why in a function such as 1/(1-x2)
linear approximation of 1/(1-u) where u = x2 is the same as quadratic approximation of 1/(1-x2)

The linear approximation is 1+u or 1+x2
Quadratic approximation is the same, 1+x2

Can someone explain to me why this happens?

Thanks!
 
Physics news on Phys.org
  • #2
JS-Student said:
Hi

I'm having trouble visualizing why in a function such as 1/(1-x2)
linear approximation of 1/(1-u) where u = x2 is the same as quadratic approximation of 1/(1-x2)

The linear approximation is 1+u or 1+x2
Quadratic approximation is the same, 1+x2

Can someone explain to me why this happens?
If you carry out the long division for ##\frac 1 {1 - x^2}## you get ##\frac 1 {1 - x^2} = 1 + x^2 + x^4 + \dots##. The quadratic approximation (using terms of degree 2 or less) of ##\frac 1 {1 - x^2}## is just ##1 + x^2##.
Similarly, if you carry out the long division for ##\frac 1 {1 - u}## you get ##\frac 1 {1 - u} = 1 + u + u^2 + \dots##. The linearization (which uses terms of degree 1 or less) is 1 + u. If ##u = x^2##, the quadratic approximation of ##\frac 1 {1 - x^2}## is the same as the linear approximation of ##\frac 1 {1 - u}##.
 
  • Like
Likes JS-Student
  • #3
It comes from composition laws for polynomial approximations:
if two functions ##f## and ##g## have a polynomial approximation in 0 at order ##n##, say ## P## and ## Q## respectively, then if ##f(0) =0##, ## g \circ f ## has a polynomial approximation at order ## n##, which is the truncature at degree ##n## of ## Q \circ P##.

In your case, ##f(x) = x^2## admits a polynomial approximation at order 2, which is ## P(x) = x^2 ##,
as well as ##g(x) = \frac{1}{1-u}## which has polynomial approximation ##Q(x) = 1 + x + x^2 ## at order 2. Furthermore ## f(0) = 0##, so the polynomial approximation of ## g\circ f## is the truncature of ##Q\circ P = 1 + x^2 + x^4## at degree 2, which is ## 1+x^2##.
 
  • Like
Likes JS-Student

Related to Why can linear approximation equal quadratic approximation

1. Why can linear approximation equal quadratic approximation?

Linear and quadratic approximations are both methods used to estimate the behavior of a function. Linear approximation uses a straight line to approximate a function, while quadratic approximation uses a parabola. However, in some cases, the linear approximation can be equal to the quadratic approximation. This is because a parabola is essentially two straight lines put together, so it is possible for a linear approximation to approximate the function in the same way that a quadratic approximation would.

2. Can linear approximation be used for all functions?

No, linear approximation can only be used for functions that are relatively simple and have a small range of values. It is not suitable for functions with sharp curves or those that vary significantly over their domain.

3. How accurate is quadratic approximation compared to linear approximation?

Quadratic approximation is generally more accurate than linear approximation because it takes into account the curvature of a function, while linear approximation only considers the slope. However, the accuracy of both methods depends on the specific function being approximated and the interval chosen for the approximation.

4. Can linear approximation and quadratic approximation be used interchangeably?

No, linear and quadratic approximations are two distinct methods and cannot be used interchangeably. Linear approximation is used to estimate a function's behavior near a single point, while quadratic approximation is used to estimate the behavior over a small interval of the function.

5. Is it always necessary to use quadratic approximation when linear approximation is not accurate enough?

No, there are other methods for approximating functions, such as cubic approximation or polynomial approximation, that may be more accurate than both linear and quadratic approximation. The choice of approximation method depends on the function and the desired level of accuracy.

Similar threads

Replies
1
Views
1K
Replies
12
Views
972
Replies
3
Views
1K
Replies
10
Views
2K
Replies
30
Views
2K
Replies
3
Views
889
Back
Top