MHB Interpolation with cubic splines

evinda
Gold Member
MHB
Messages
3,741
Reaction score
0
Hello! :) I am looking at the proof of this theorem:
Let $f \in C^{1}([a,b]),P:a=x_{0}<x_{1}<...<x_{n}=b $ uniform partition of $[a,b]$.Then there is exactly one function $s \in S_{3}(P)$ so that $s(x_{i})=f(x_{i}),i=0,...,n$ and $s,s',s''$ continuous at $x_{i}$.Also,$s'(x_{0})=f'(x_{0}),s'(x_{n})=f'(x_{n})$.

but at some point I got stuck :confused:

Let $s'_{i}=s'(x_{i}),s''_{i}=s''(x_{i}),s_{i}=s(x_{i}),y_{i}=f(x_{i})$

From the linear Lagrange polynomial,we find that:
$$s''(x)=\frac{1}{h}[s''_{i}(x-x_{i-1})-s''_{i-1}(x-x_{i})] ,x \in [x_{i-1},x_{i}]$$

Then,integrating two times,we get:
$$s(x)=\frac{1}{6h}[s''_{i}(x-x_{i-1})^{3}-s''_{i-1}(x-x_{i})^{3}]+c_{i1}x+c_{i0}$$
Then using the conditions $s_{i-1}=y_{i-1},s_{i}=y_{i}$, we want to find the constants $c_{i1},c_{i0}$ .In my textbook,there is only the solution,which is $s(x)=\frac{1}{6h}[s''_{i}(x-x_{i-1})^{3}-s''_{i-1}(x-x_{i})^{3}]+(\frac{y_{i}}{h}-s''_{i}\frac{h}{6})(x-x_{i-1})-(\frac{y_{i-1}}{h}-s''_{i-1}\frac{h}{6})(x-x_{i})$

I tried to find the constants..For $c_{i1}$ I found $\frac{y_{i}-y_{i-1}}{h}-\frac{h}{6}(s''_{i}-s''_{i-1})$ ,but replacing the conditions and $c_{i1}$, I don't get the right result! Have I calculated wrong the value of $c_{i1}$ ?
From the condition $s_{i-1}=y_{i-1} \text{ and } c_{i1}$ I get $$c_{i0}=y_{i-1}-\frac{h^{2}}{6}s''_{i-1}-\frac{y_{i}-y_{i-1}}{h}x_{i-1}+\frac{h}{6}(s''_{i}-s''_{i-1})x_{i-1}$$
and from the condition $s_{i}=y_{i} \text{ and } c_{i1}$ I get $$c_{i0}=y_{i}-\frac{h^{2}}{6}s''_{i}-\frac{y_{i}-y_{i-1}}{h}x_{i}+\frac{h}{6}(s''_{i}-s''_{i-1})x_{i}$$
 
Mathematics news on Phys.org
Hey evinda! :D

evinda said:
Hello! :) I am looking at the proof of this theorem:
Let $f \in C^{1}([a,b]),P:a=x_{0}<x_{1}<...<x_{n}=b $ uniform partition of $[a,b]$.Then there is exactly one function $s \in S_{3}(P)$ so that $s(x_{i})=f(x_{i}),i=0,...,n$ and $s,s',s''$ continuous at $x_{i}$.Also,$s'(x_{0})=f'(x_{0}),s'(x_{n})=f'(x_{n})$.

but at some point I got stuck :confused:

Let $s'_{i}=s'(x_{i}),s''_{i}=s''(x_{i}),s_{i}=s(x_{i}),y_{i}=f(x_{i})$

From the linear Lagrange polynomial,we find that:
$$s''(x)=\frac{1}{h}[s''_{i}(x-x_{i-1})-s''_{i-1}(x-x_{i})] ,x \in [x_{i-1},x_{i}]$$

Then,integrating two times,we get:
$$s(x)=\frac{1}{6h}[s''_{i}(x-x_{i-1})^{3}-s''_{i-1}(x-x_{i})^{3}]+c_{i1}x+c_{i0}$$
Then using the conditions $s_{i-1}=y_{i-1},s_{i}=y_{i}$, we want to find the constants $c_{i1},c_{i0}$ .In my textbook,there is only the solution,which is $s(x)=\frac{1}{6h}[s''_{i}(x-x_{i-1})^{3}-s''_{i-1}(x-x_{i})^{3}]+(\frac{y_{i}}{h}-s''_{i}\frac{h}{6})(x-x_{i-1})-(\frac{y_{i-1}}{h}-s''_{i-1}\frac{h}{6})(x-x_{i})$

I tried to find the constants..For $c_{i1}$ I found $\frac{y_{i}-y_{i-1}}{h}-\frac{h}{6}(s''_{i}-s''_{i-1})$ ,but replacing the conditions and $c_{i1}$, I don't get the right result! Have I calculated wrong the value of $c_{i1}$ ?

Nope. Your value of $c_{i1}$ is all good. (Nod)
Note that the given solution has factors $(x-x_i)$ and $(x-x_{i-1})$.
If you bundle the factors of $x$ from the solution, you get your $c_{i1}$.
From the condition $s_{i-1}=y_{i-1} \text{ and } c_{i1}$ I get $$c_{i0}=y_{i-1}-\frac{h^{2}}{6}s''_{i-1}-\frac{y_{i}-y_{i-1}}{h}x_{i-1}+\frac{h}{6}(s''_{i}-s''_{i-1})x_{i-1}$$
and from the condition $s_{i}=y_{i} \text{ and } c_{i1}$ I get $$c_{i0}=y_{i}-\frac{h^{2}}{6}s''_{i}-\frac{y_{i}-y_{i-1}}{h}x_{i}+\frac{h}{6}(s''_{i}-s''_{i-1})x_{i}$$

Well... I guess the solution should be of the form:
$$... + C(x-x_i) + D(x-x_{i-1}) + E$$
instead of
$$... + c_{i1}x+c_{i0}$$
I haven't checked, but presumably your $c_{i0}$ will be equal to $-Cx_i - Dx_{i-1}+E$.
 
Insights auto threads is broken atm, so I'm manually creating these for new Insight articles. In Dirac’s Principles of Quantum Mechanics published in 1930 he introduced a “convenient notation” he referred to as a “delta function” which he treated as a continuum analog to the discrete Kronecker delta. The Kronecker delta is simply the indexed components of the identity operator in matrix algebra Source: https://www.physicsforums.com/insights/what-exactly-is-diracs-delta-function/ by...
Fermat's Last Theorem has long been one of the most famous mathematical problems, and is now one of the most famous theorems. It simply states that the equation $$ a^n+b^n=c^n $$ has no solutions with positive integers if ##n>2.## It was named after Pierre de Fermat (1607-1665). The problem itself stems from the book Arithmetica by Diophantus of Alexandria. It gained popularity because Fermat noted in his copy "Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos, et...
I'm interested to know whether the equation $$1 = 2 - \frac{1}{2 - \frac{1}{2 - \cdots}}$$ is true or not. It can be shown easily that if the continued fraction converges, it cannot converge to anything else than 1. It seems that if the continued fraction converges, the convergence is very slow. The apparent slowness of the convergence makes it difficult to estimate the presence of true convergence numerically. At the moment I don't know whether this converges or not.
Back
Top