MHB Inequality: $$-\frac{22}{24}h^2 ||f^{(4)}||_{\infty}\leq 0$$

  • Thread starter Thread starter mathmari
  • Start date Start date
Click For Summary
The discussion centers on the validity of an inequality involving the fourth derivative of a function, specifically whether the expression $$|-\frac{5}{24}h^2f^{(4)}(x_1)+\frac{64}{24}h^2f^{(4)}(x_2)-\frac{81}{24}h^2f^{(4)}(x_3}|$$ can be bounded as proposed. It is argued that the inequality does not hold unless the fourth derivatives at the specified points share the same sign, as differing signs could lead to a larger absolute value than suggested. The Taylor expansion is utilized to derive the relationship, leading to a conclusion that the best bound achievable is $$\frac{150}{24}h^2\|f^{(4)}\|_{\infty}$$ instead of the initially proposed bound. The final goal is to show that the difference between the finite difference approximation and the second derivative is bounded by $$\frac{11}{12} h^2 ||f^{(4)}||_{\infty}$$, which is affirmed to be correct under the given assumptions.
mathmari
Gold Member
MHB
Messages
4,984
Reaction score
7
Hey! :o

Does the following inequality hold?

$$|-\frac{5}{24}h^2f^{(4)}(x_1)+\frac{64}{24}h^2f^{(4)}(x_2)-\frac{81}{24}h^2f^{(4)}(x_3)| \\ \leq |-\frac{5}{24}h^2+\frac{64}{24}h^2-\frac{81}{24}h^2|\max_x |f^{(4)}(x)| \\ = |-\frac{22}{24}h^2| ||f^{(4)}||_{\infty}= \frac{22}{24}h^2 ||f^{(4)}||_{\infty}$$

where $h>0$
 
Physics news on Phys.org
mathmari said:
Hey! :o

Does the following inequality hold?

$$|-\frac{5}{24}h^2f^{(4)}(x_1)+\frac{64}{24}h^2f^{(4)}(x_2)-\frac{81}{24}h^2f^{(4)}(x_3)| \\ \leq |-\frac{5}{24}h^2+\frac{64}{24}h^2-\frac{81}{24}h^2|\max_x |f^{(4)}(x)| \\ = |-\frac{22}{24}h^2| ||f^{(4)}||_{\infty}= \frac{22}{24}h^2 ||f^{(4)}||_{\infty}$$

where $h>0$
No, because $f^{(4)}(x_2)$ might have a different sign from $f^{(4)}(x_1)$ and $f^{(4)}(x_3)$. Unless you know that $f^{(4)}(x_1)$, $f^{(4)}(x_2)$ and $f^{(4)}(x_3)$ all have the same sign, the best you can say is that $$\left|-\frac{5}{24}h^2f^{(4)}(x_1)+\frac{64}{24}h^2f^{(4)}(x_2)-\frac{81}{24}h^2f^{(4)}(x_3)\right| \leqslant \left(\frac5{24} + \frac{64}{24} + \frac{81}{24}\right)h^2\|f^{(4)}\|_{\infty} = \frac{150}{24}h^2\|f^{(4)}\|_{\infty}.$$
 
I used the Taylor expansion three times, once for $f(x+h)$, once for $f(x+2h)$ and once for $f(x+3h)$ and the $f(x_1), f(x_2), f(x_3)$ of the above expression are the remainders of each expansion. Do we know if they have all the same sign?
 
I want to show that $$|\delta_{h,r}f(x)- f''(x)| \leq \frac{11}{12} h^2 ||f^{(4)}||_{\infty}$$ where $$\delta_{h,r}f(x)=\frac{1}{h^2} (2f(x)-5f(x+h)+4f(x+2h)-f(x+3h))$$

I applied the Taylor expanson at $f(x+h)$, $f(x+2h)$ and $f(x+3h)$ and $f(x_1), f(x_2), f(x_3)$ are the corresponding remainder of each expansion.

$$f(x+h)=f(x)+h f'(x)+\frac{h^2}{2} f''(x)+\frac{h^3}{6} f'''(x)+\frac{h^4}{24} f^{(4)}(x_1), x_1 \in (x,x+h)$$

$$f(x+2h)=f(x)+2hf'(x)+2h^2 f''(x)+\frac{4}{3} h^3 f'''(x)+\frac{16}{24}f^{(4)}(x_2), x_2 \in (x,x+2h)$$

$$f(x+3h)=f(x)+3hf'(x)+\frac{9}{2}h^2 f''(x)+\frac{27}{6}h^3 f'''(x)+\frac{81}{24} h^4 f^{(4)}(x_3), x_3 \in (x,x+3h)$$Substituting at $$\delta_{h,r}f(x)=\frac{1}{h^2} (2f(x)-5f(x+h)+4f(x+2h)-f(x+3h))$$ we get
$$\delta_{h,r}=f''(x)+\frac{h^2}{24}(64 f^{(4)}(x_2)-5f^{(4)}(x_1)-81 f^{(4)}(x_3))$$

Since $$|-\frac{5}{24}h^2f^{(4)}(x_1)+\frac{64}{24}h^2f^{(4)}(x_2)-\frac{81}{24}h^2f^{(4)}(x_3)| \\ \leq |-\frac{5}{24}h^2+\frac{64}{24}h^2-\frac{81}{24}h^2|\max_x |f^{(4)}(x)| \\ = |-\frac{22}{24}h^2| ||f^{(4)}||_{\infty}= \frac{22}{24}h^2 ||f^{(4)}||_{\infty}$$
we have

$$|\delta_{h,r}f(x)- f''(x)| \leq \frac{11}{12} h^2 ||f^{(4)}||_{\infty}$$

($h>0$)

Is this correct? Is there somewhere a mistake?
 
Thread 'Problem with calculating projections of curl using rotation of contour'
Hello! I tried to calculate projections of curl using rotation of coordinate system but I encountered with following problem. Given: ##rot_xA=\frac{\partial A_z}{\partial y}-\frac{\partial A_y}{\partial z}=0## ##rot_yA=\frac{\partial A_x}{\partial z}-\frac{\partial A_z}{\partial x}=1## ##rot_zA=\frac{\partial A_y}{\partial x}-\frac{\partial A_x}{\partial y}=0## I rotated ##yz##-plane of this coordinate system by an angle ##45## degrees about ##x##-axis and used rotation matrix to...

Similar threads

  • · Replies 16 ·
Replies
16
Views
4K
  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 24 ·
Replies
24
Views
3K
Replies
7
Views
2K
Replies
18
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K