Finding an interpolating polynomial based upon four points
$begingroup$
Below is a problem I made up and my incorrect solution to it. I am hoping that somebody here can tell me where I went wrong.
Thanks,
Bob
Problem:
Given the points $(0,0), (1,3), (2,5) (4,4)$ find a second order
interpolating polynomial, of the form $f(x) = Ax^2 + Bx + C$, such that the point $(4,4)$ is on the curve and the following is minimized:
$$ (f(0)- 0)^2 + (f(1) -3)^2 + (f(2) - 5)^2 $$
Answer:
In my answer, I write $D_A$ for the partial derivative of $D$ with $A$. $D_B$ has a similar meaning.
begin{align*}
f(4) &= A(4^2) + B(4) + C = 4 \
16A + 4B + C &= 4 \
C &= 4 - 16A - 4B \
end{align*}
Now let $D = (f(0)- 0)^2 + (f(1) -3)^2 + (f(2) - 5)^2$.
begin{align*}
(f(0)- 0)^2 &= C^2 = (4 - 16A - 4B )^2 \
(f(1) -3)^2 &= (A + B + C - 3)^2 = (A + B + 4 - 16A - 4B - 3)^2 \
(f(1) -3)^2 &= (-15A -3B + 1)^2 \
(f(2) - 5)^2 &= (4A + 2B + C - 5)^2 = (4A + 2B + 4 - 16A - 4B - 5)^2 \
(f(2) - 5)^2 &= (-12A -2B - 1)^2 \
D &= (4 - 16A - 4B )^2 + (-15A -3B + 1)^2 + (-12A -2B - 1)^2 \
D_A &= -32( 4 - 16A - 4B) - 30( -15A - 3B + 1 ) - 24(-12A - 2B - 1) \
D_A &= -128 + 16(32)A + 128B + 30(15)A + 90B - 30 + 24(12)A + 48B + 24 \
D_A &= 512A + 128B + 450A + 90B + 288A + 48B - 134 \
D_A &= 1250A + 266B - 134 \
end{align*}
begin{align*}
D_B &= -8( 4 - 16A - 4B ) - 6( -15A - 3B + 1 ) - 4( -12A - 2B - 1 ) \
D_B &= -32 +8(16)A + 32B + 90A + 18B - 6 + 48A + 8B + 4 \
D_B &= 128A + 32B + 90A + 18B + 48A + 8B - 34 \
D_B &= 266A + 98B - 34 \
end{align*}
begin{align*}
1250A + 266B - 134 &= 0 \
266A + 98B - 34 &= 0 \
133A + 49B - 17 &= 0 \
A &= frac{17-49B}{133} \
625A + 133B - 67 &= 0 \
frac{625(17-49B)}{133} + 133B &= 67 \
frac{10625 - 30625B}{133} + 133B &= 67 \
10625 - 30625B + 17689B &= 8911 \
10625 - 12936B &= 8911 \
B &= -frac{1714}{12936} \
end{align*}
begin{align*}
A &= frac{17 - frac{-49(1714)}{12936} } {133} = frac{17 + frac{83986}{12936} }{133} \
A &= frac{17 + frac{41993}{6468} }{133} \
A &= frac{ 17(12936) + 83986 } { 12936(133)} = frac{303898} {1720488} \
A &= frac{ 6202}{ 35112 } = frac{ 3101 }{17556} \
A &= frac{ 443}{2508} \
C &= 4 - 16left( frac{ 443}{2508} right) - 4 left( -frac{1714}{12936} right) \
C &= 4 - 16left( frac{ 443}{2508} right) + 4 left( frac{1714}{12936} right) \
C &= 4 - frac{4(443)}{627} + frac{1714}{3234} \
C &= 4 - frac{1772}{627} + frac{1714}{3234} \
C &= frac{4(627)(3234) - 1772(3234) + 1714(627)} {627(3234)} \
C &= frac{ 3454902} {2027718 } = frac{191939} {112651} \
C &= frac{ 17449 }{ 10241 } \
end{align*}
Using the program R, I find that:
$$ f(x) = -0.72936 x^2 + 3.93119 x - 0.05505 $$
Hence, I conclude my answer must be wrong.
statistics multivariable-calculus
$endgroup$
add a comment |
$begingroup$
Below is a problem I made up and my incorrect solution to it. I am hoping that somebody here can tell me where I went wrong.
Thanks,
Bob
Problem:
Given the points $(0,0), (1,3), (2,5) (4,4)$ find a second order
interpolating polynomial, of the form $f(x) = Ax^2 + Bx + C$, such that the point $(4,4)$ is on the curve and the following is minimized:
$$ (f(0)- 0)^2 + (f(1) -3)^2 + (f(2) - 5)^2 $$
Answer:
In my answer, I write $D_A$ for the partial derivative of $D$ with $A$. $D_B$ has a similar meaning.
begin{align*}
f(4) &= A(4^2) + B(4) + C = 4 \
16A + 4B + C &= 4 \
C &= 4 - 16A - 4B \
end{align*}
Now let $D = (f(0)- 0)^2 + (f(1) -3)^2 + (f(2) - 5)^2$.
begin{align*}
(f(0)- 0)^2 &= C^2 = (4 - 16A - 4B )^2 \
(f(1) -3)^2 &= (A + B + C - 3)^2 = (A + B + 4 - 16A - 4B - 3)^2 \
(f(1) -3)^2 &= (-15A -3B + 1)^2 \
(f(2) - 5)^2 &= (4A + 2B + C - 5)^2 = (4A + 2B + 4 - 16A - 4B - 5)^2 \
(f(2) - 5)^2 &= (-12A -2B - 1)^2 \
D &= (4 - 16A - 4B )^2 + (-15A -3B + 1)^2 + (-12A -2B - 1)^2 \
D_A &= -32( 4 - 16A - 4B) - 30( -15A - 3B + 1 ) - 24(-12A - 2B - 1) \
D_A &= -128 + 16(32)A + 128B + 30(15)A + 90B - 30 + 24(12)A + 48B + 24 \
D_A &= 512A + 128B + 450A + 90B + 288A + 48B - 134 \
D_A &= 1250A + 266B - 134 \
end{align*}
begin{align*}
D_B &= -8( 4 - 16A - 4B ) - 6( -15A - 3B + 1 ) - 4( -12A - 2B - 1 ) \
D_B &= -32 +8(16)A + 32B + 90A + 18B - 6 + 48A + 8B + 4 \
D_B &= 128A + 32B + 90A + 18B + 48A + 8B - 34 \
D_B &= 266A + 98B - 34 \
end{align*}
begin{align*}
1250A + 266B - 134 &= 0 \
266A + 98B - 34 &= 0 \
133A + 49B - 17 &= 0 \
A &= frac{17-49B}{133} \
625A + 133B - 67 &= 0 \
frac{625(17-49B)}{133} + 133B &= 67 \
frac{10625 - 30625B}{133} + 133B &= 67 \
10625 - 30625B + 17689B &= 8911 \
10625 - 12936B &= 8911 \
B &= -frac{1714}{12936} \
end{align*}
begin{align*}
A &= frac{17 - frac{-49(1714)}{12936} } {133} = frac{17 + frac{83986}{12936} }{133} \
A &= frac{17 + frac{41993}{6468} }{133} \
A &= frac{ 17(12936) + 83986 } { 12936(133)} = frac{303898} {1720488} \
A &= frac{ 6202}{ 35112 } = frac{ 3101 }{17556} \
A &= frac{ 443}{2508} \
C &= 4 - 16left( frac{ 443}{2508} right) - 4 left( -frac{1714}{12936} right) \
C &= 4 - 16left( frac{ 443}{2508} right) + 4 left( frac{1714}{12936} right) \
C &= 4 - frac{4(443)}{627} + frac{1714}{3234} \
C &= 4 - frac{1772}{627} + frac{1714}{3234} \
C &= frac{4(627)(3234) - 1772(3234) + 1714(627)} {627(3234)} \
C &= frac{ 3454902} {2027718 } = frac{191939} {112651} \
C &= frac{ 17449 }{ 10241 } \
end{align*}
Using the program R, I find that:
$$ f(x) = -0.72936 x^2 + 3.93119 x - 0.05505 $$
Hence, I conclude my answer must be wrong.
statistics multivariable-calculus
$endgroup$
1
$begingroup$
Error: sign of B is wrong, should be +.
$endgroup$
– herb steinberg
Jan 6 at 22:54
$begingroup$
@herbsteinberg I agree that I have a sign error for $B$. However, I still get a wrong value for $B$.
$endgroup$
– Bob
Jan 6 at 23:14
$begingroup$
Did you recalculate A and C, both of which use the value of B? B$=+frac{1714}{12936}$.
$endgroup$
– herb steinberg
Jan 7 at 2:48
add a comment |
$begingroup$
Below is a problem I made up and my incorrect solution to it. I am hoping that somebody here can tell me where I went wrong.
Thanks,
Bob
Problem:
Given the points $(0,0), (1,3), (2,5) (4,4)$ find a second order
interpolating polynomial, of the form $f(x) = Ax^2 + Bx + C$, such that the point $(4,4)$ is on the curve and the following is minimized:
$$ (f(0)- 0)^2 + (f(1) -3)^2 + (f(2) - 5)^2 $$
Answer:
In my answer, I write $D_A$ for the partial derivative of $D$ with $A$. $D_B$ has a similar meaning.
begin{align*}
f(4) &= A(4^2) + B(4) + C = 4 \
16A + 4B + C &= 4 \
C &= 4 - 16A - 4B \
end{align*}
Now let $D = (f(0)- 0)^2 + (f(1) -3)^2 + (f(2) - 5)^2$.
begin{align*}
(f(0)- 0)^2 &= C^2 = (4 - 16A - 4B )^2 \
(f(1) -3)^2 &= (A + B + C - 3)^2 = (A + B + 4 - 16A - 4B - 3)^2 \
(f(1) -3)^2 &= (-15A -3B + 1)^2 \
(f(2) - 5)^2 &= (4A + 2B + C - 5)^2 = (4A + 2B + 4 - 16A - 4B - 5)^2 \
(f(2) - 5)^2 &= (-12A -2B - 1)^2 \
D &= (4 - 16A - 4B )^2 + (-15A -3B + 1)^2 + (-12A -2B - 1)^2 \
D_A &= -32( 4 - 16A - 4B) - 30( -15A - 3B + 1 ) - 24(-12A - 2B - 1) \
D_A &= -128 + 16(32)A + 128B + 30(15)A + 90B - 30 + 24(12)A + 48B + 24 \
D_A &= 512A + 128B + 450A + 90B + 288A + 48B - 134 \
D_A &= 1250A + 266B - 134 \
end{align*}
begin{align*}
D_B &= -8( 4 - 16A - 4B ) - 6( -15A - 3B + 1 ) - 4( -12A - 2B - 1 ) \
D_B &= -32 +8(16)A + 32B + 90A + 18B - 6 + 48A + 8B + 4 \
D_B &= 128A + 32B + 90A + 18B + 48A + 8B - 34 \
D_B &= 266A + 98B - 34 \
end{align*}
begin{align*}
1250A + 266B - 134 &= 0 \
266A + 98B - 34 &= 0 \
133A + 49B - 17 &= 0 \
A &= frac{17-49B}{133} \
625A + 133B - 67 &= 0 \
frac{625(17-49B)}{133} + 133B &= 67 \
frac{10625 - 30625B}{133} + 133B &= 67 \
10625 - 30625B + 17689B &= 8911 \
10625 - 12936B &= 8911 \
B &= -frac{1714}{12936} \
end{align*}
begin{align*}
A &= frac{17 - frac{-49(1714)}{12936} } {133} = frac{17 + frac{83986}{12936} }{133} \
A &= frac{17 + frac{41993}{6468} }{133} \
A &= frac{ 17(12936) + 83986 } { 12936(133)} = frac{303898} {1720488} \
A &= frac{ 6202}{ 35112 } = frac{ 3101 }{17556} \
A &= frac{ 443}{2508} \
C &= 4 - 16left( frac{ 443}{2508} right) - 4 left( -frac{1714}{12936} right) \
C &= 4 - 16left( frac{ 443}{2508} right) + 4 left( frac{1714}{12936} right) \
C &= 4 - frac{4(443)}{627} + frac{1714}{3234} \
C &= 4 - frac{1772}{627} + frac{1714}{3234} \
C &= frac{4(627)(3234) - 1772(3234) + 1714(627)} {627(3234)} \
C &= frac{ 3454902} {2027718 } = frac{191939} {112651} \
C &= frac{ 17449 }{ 10241 } \
end{align*}
Using the program R, I find that:
$$ f(x) = -0.72936 x^2 + 3.93119 x - 0.05505 $$
Hence, I conclude my answer must be wrong.
statistics multivariable-calculus
$endgroup$
Below is a problem I made up and my incorrect solution to it. I am hoping that somebody here can tell me where I went wrong.
Thanks,
Bob
Problem:
Given the points $(0,0), (1,3), (2,5) (4,4)$ find a second order
interpolating polynomial, of the form $f(x) = Ax^2 + Bx + C$, such that the point $(4,4)$ is on the curve and the following is minimized:
$$ (f(0)- 0)^2 + (f(1) -3)^2 + (f(2) - 5)^2 $$
Answer:
In my answer, I write $D_A$ for the partial derivative of $D$ with $A$. $D_B$ has a similar meaning.
begin{align*}
f(4) &= A(4^2) + B(4) + C = 4 \
16A + 4B + C &= 4 \
C &= 4 - 16A - 4B \
end{align*}
Now let $D = (f(0)- 0)^2 + (f(1) -3)^2 + (f(2) - 5)^2$.
begin{align*}
(f(0)- 0)^2 &= C^2 = (4 - 16A - 4B )^2 \
(f(1) -3)^2 &= (A + B + C - 3)^2 = (A + B + 4 - 16A - 4B - 3)^2 \
(f(1) -3)^2 &= (-15A -3B + 1)^2 \
(f(2) - 5)^2 &= (4A + 2B + C - 5)^2 = (4A + 2B + 4 - 16A - 4B - 5)^2 \
(f(2) - 5)^2 &= (-12A -2B - 1)^2 \
D &= (4 - 16A - 4B )^2 + (-15A -3B + 1)^2 + (-12A -2B - 1)^2 \
D_A &= -32( 4 - 16A - 4B) - 30( -15A - 3B + 1 ) - 24(-12A - 2B - 1) \
D_A &= -128 + 16(32)A + 128B + 30(15)A + 90B - 30 + 24(12)A + 48B + 24 \
D_A &= 512A + 128B + 450A + 90B + 288A + 48B - 134 \
D_A &= 1250A + 266B - 134 \
end{align*}
begin{align*}
D_B &= -8( 4 - 16A - 4B ) - 6( -15A - 3B + 1 ) - 4( -12A - 2B - 1 ) \
D_B &= -32 +8(16)A + 32B + 90A + 18B - 6 + 48A + 8B + 4 \
D_B &= 128A + 32B + 90A + 18B + 48A + 8B - 34 \
D_B &= 266A + 98B - 34 \
end{align*}
begin{align*}
1250A + 266B - 134 &= 0 \
266A + 98B - 34 &= 0 \
133A + 49B - 17 &= 0 \
A &= frac{17-49B}{133} \
625A + 133B - 67 &= 0 \
frac{625(17-49B)}{133} + 133B &= 67 \
frac{10625 - 30625B}{133} + 133B &= 67 \
10625 - 30625B + 17689B &= 8911 \
10625 - 12936B &= 8911 \
B &= -frac{1714}{12936} \
end{align*}
begin{align*}
A &= frac{17 - frac{-49(1714)}{12936} } {133} = frac{17 + frac{83986}{12936} }{133} \
A &= frac{17 + frac{41993}{6468} }{133} \
A &= frac{ 17(12936) + 83986 } { 12936(133)} = frac{303898} {1720488} \
A &= frac{ 6202}{ 35112 } = frac{ 3101 }{17556} \
A &= frac{ 443}{2508} \
C &= 4 - 16left( frac{ 443}{2508} right) - 4 left( -frac{1714}{12936} right) \
C &= 4 - 16left( frac{ 443}{2508} right) + 4 left( frac{1714}{12936} right) \
C &= 4 - frac{4(443)}{627} + frac{1714}{3234} \
C &= 4 - frac{1772}{627} + frac{1714}{3234} \
C &= frac{4(627)(3234) - 1772(3234) + 1714(627)} {627(3234)} \
C &= frac{ 3454902} {2027718 } = frac{191939} {112651} \
C &= frac{ 17449 }{ 10241 } \
end{align*}
Using the program R, I find that:
$$ f(x) = -0.72936 x^2 + 3.93119 x - 0.05505 $$
Hence, I conclude my answer must be wrong.
statistics multivariable-calculus
statistics multivariable-calculus
asked Jan 6 at 22:12
BobBob
914514
914514
1
$begingroup$
Error: sign of B is wrong, should be +.
$endgroup$
– herb steinberg
Jan 6 at 22:54
$begingroup$
@herbsteinberg I agree that I have a sign error for $B$. However, I still get a wrong value for $B$.
$endgroup$
– Bob
Jan 6 at 23:14
$begingroup$
Did you recalculate A and C, both of which use the value of B? B$=+frac{1714}{12936}$.
$endgroup$
– herb steinberg
Jan 7 at 2:48
add a comment |
1
$begingroup$
Error: sign of B is wrong, should be +.
$endgroup$
– herb steinberg
Jan 6 at 22:54
$begingroup$
@herbsteinberg I agree that I have a sign error for $B$. However, I still get a wrong value for $B$.
$endgroup$
– Bob
Jan 6 at 23:14
$begingroup$
Did you recalculate A and C, both of which use the value of B? B$=+frac{1714}{12936}$.
$endgroup$
– herb steinberg
Jan 7 at 2:48
1
1
$begingroup$
Error: sign of B is wrong, should be +.
$endgroup$
– herb steinberg
Jan 6 at 22:54
$begingroup$
Error: sign of B is wrong, should be +.
$endgroup$
– herb steinberg
Jan 6 at 22:54
$begingroup$
@herbsteinberg I agree that I have a sign error for $B$. However, I still get a wrong value for $B$.
$endgroup$
– Bob
Jan 6 at 23:14
$begingroup$
@herbsteinberg I agree that I have a sign error for $B$. However, I still get a wrong value for $B$.
$endgroup$
– Bob
Jan 6 at 23:14
$begingroup$
Did you recalculate A and C, both of which use the value of B? B$=+frac{1714}{12936}$.
$endgroup$
– herb steinberg
Jan 7 at 2:48
$begingroup$
Did you recalculate A and C, both of which use the value of B? B$=+frac{1714}{12936}$.
$endgroup$
– herb steinberg
Jan 7 at 2:48
add a comment |
1 Answer
1
active
oldest
votes
$begingroup$
I would do this differently, given you have a point on the curve.
First set $f(x)=A(x-4)(x-E)+4$ to give the point on the curve and to reduce the number of variables. We will put $AE=F$, since $E$ only appears as a product with $A$
Then
$f(0)=4F+4$,
$f(1)-3=-3A(1-E)+4-3=3F+1-3A$
$f(2)-5=-2A(2-E)+4-5=2F+4-4A$
Now I notice that I can simplify further if I set $G=F-A$ so that I can write the expression I want to minimise as $$16(F+1)^2+(3G+1)^2+4(2G-F+2)^2$$
If I fix $F$ and minimise with respect to $G$ I find I need $$6(3G+1)+8(2G-F+2)=0=34G-16F+22$$ so that $$G=frac {8F-11}{17}$$
and I continue by substituting back for $G$, and then work back from $G$ and $F$ to $A$ and $E$, having simplified the computation somewhat.
$endgroup$
$begingroup$
You wrote: "If I fix F and minimise with respect to G I find I need" I do not understand why that is a valid thing to do. Are you taking a partial derivative?
$endgroup$
– Bob
Jan 6 at 23:36
$begingroup$
I am thinking that you have only one equation and you need a second one. That is, you have two unknowns.
$endgroup$
– Bob
Jan 6 at 23:41
1
$begingroup$
@Bob The transformations I am making are reversible, so that a minimum value in respect of one set of parameters is also a minimum in respect of another. This is easy here because I am minimising a positive definite quadratic form. In terms of the equations, the point given on the curve gives one relation between the coefficients and reduces the number of necessary parameters to two. I have chosen to reduce and simplify before doing any other calculations. Your method keeps $A, B, C$
$endgroup$
– Mark Bennet
Jan 7 at 7:01
1
$begingroup$
@Bob Yes, I am taking partial derivative with respect to $G$. Since the form here is positive definite, it is easy to see that any local extremum is a minimum. Since it is linearly parametrised by two parameters ($A, F$ or $F, G$) it turns out that there is a single local minimum which is a global minimum. I also need to take the partial derivative with respect to $F$ to finish (or I can complete the square, which comes to the same thing). Since there is a single local minimum we can (with just a little care) locate it using the variables we choose rather than those we started with.
$endgroup$
– Mark Bennet
Jan 7 at 7:06
add a comment |
Your Answer
StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "69"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3064435%2ffinding-an-interpolating-polynomial-based-upon-four-points%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
$begingroup$
I would do this differently, given you have a point on the curve.
First set $f(x)=A(x-4)(x-E)+4$ to give the point on the curve and to reduce the number of variables. We will put $AE=F$, since $E$ only appears as a product with $A$
Then
$f(0)=4F+4$,
$f(1)-3=-3A(1-E)+4-3=3F+1-3A$
$f(2)-5=-2A(2-E)+4-5=2F+4-4A$
Now I notice that I can simplify further if I set $G=F-A$ so that I can write the expression I want to minimise as $$16(F+1)^2+(3G+1)^2+4(2G-F+2)^2$$
If I fix $F$ and minimise with respect to $G$ I find I need $$6(3G+1)+8(2G-F+2)=0=34G-16F+22$$ so that $$G=frac {8F-11}{17}$$
and I continue by substituting back for $G$, and then work back from $G$ and $F$ to $A$ and $E$, having simplified the computation somewhat.
$endgroup$
$begingroup$
You wrote: "If I fix F and minimise with respect to G I find I need" I do not understand why that is a valid thing to do. Are you taking a partial derivative?
$endgroup$
– Bob
Jan 6 at 23:36
$begingroup$
I am thinking that you have only one equation and you need a second one. That is, you have two unknowns.
$endgroup$
– Bob
Jan 6 at 23:41
1
$begingroup$
@Bob The transformations I am making are reversible, so that a minimum value in respect of one set of parameters is also a minimum in respect of another. This is easy here because I am minimising a positive definite quadratic form. In terms of the equations, the point given on the curve gives one relation between the coefficients and reduces the number of necessary parameters to two. I have chosen to reduce and simplify before doing any other calculations. Your method keeps $A, B, C$
$endgroup$
– Mark Bennet
Jan 7 at 7:01
1
$begingroup$
@Bob Yes, I am taking partial derivative with respect to $G$. Since the form here is positive definite, it is easy to see that any local extremum is a minimum. Since it is linearly parametrised by two parameters ($A, F$ or $F, G$) it turns out that there is a single local minimum which is a global minimum. I also need to take the partial derivative with respect to $F$ to finish (or I can complete the square, which comes to the same thing). Since there is a single local minimum we can (with just a little care) locate it using the variables we choose rather than those we started with.
$endgroup$
– Mark Bennet
Jan 7 at 7:06
add a comment |
$begingroup$
I would do this differently, given you have a point on the curve.
First set $f(x)=A(x-4)(x-E)+4$ to give the point on the curve and to reduce the number of variables. We will put $AE=F$, since $E$ only appears as a product with $A$
Then
$f(0)=4F+4$,
$f(1)-3=-3A(1-E)+4-3=3F+1-3A$
$f(2)-5=-2A(2-E)+4-5=2F+4-4A$
Now I notice that I can simplify further if I set $G=F-A$ so that I can write the expression I want to minimise as $$16(F+1)^2+(3G+1)^2+4(2G-F+2)^2$$
If I fix $F$ and minimise with respect to $G$ I find I need $$6(3G+1)+8(2G-F+2)=0=34G-16F+22$$ so that $$G=frac {8F-11}{17}$$
and I continue by substituting back for $G$, and then work back from $G$ and $F$ to $A$ and $E$, having simplified the computation somewhat.
$endgroup$
$begingroup$
You wrote: "If I fix F and minimise with respect to G I find I need" I do not understand why that is a valid thing to do. Are you taking a partial derivative?
$endgroup$
– Bob
Jan 6 at 23:36
$begingroup$
I am thinking that you have only one equation and you need a second one. That is, you have two unknowns.
$endgroup$
– Bob
Jan 6 at 23:41
1
$begingroup$
@Bob The transformations I am making are reversible, so that a minimum value in respect of one set of parameters is also a minimum in respect of another. This is easy here because I am minimising a positive definite quadratic form. In terms of the equations, the point given on the curve gives one relation between the coefficients and reduces the number of necessary parameters to two. I have chosen to reduce and simplify before doing any other calculations. Your method keeps $A, B, C$
$endgroup$
– Mark Bennet
Jan 7 at 7:01
1
$begingroup$
@Bob Yes, I am taking partial derivative with respect to $G$. Since the form here is positive definite, it is easy to see that any local extremum is a minimum. Since it is linearly parametrised by two parameters ($A, F$ or $F, G$) it turns out that there is a single local minimum which is a global minimum. I also need to take the partial derivative with respect to $F$ to finish (or I can complete the square, which comes to the same thing). Since there is a single local minimum we can (with just a little care) locate it using the variables we choose rather than those we started with.
$endgroup$
– Mark Bennet
Jan 7 at 7:06
add a comment |
$begingroup$
I would do this differently, given you have a point on the curve.
First set $f(x)=A(x-4)(x-E)+4$ to give the point on the curve and to reduce the number of variables. We will put $AE=F$, since $E$ only appears as a product with $A$
Then
$f(0)=4F+4$,
$f(1)-3=-3A(1-E)+4-3=3F+1-3A$
$f(2)-5=-2A(2-E)+4-5=2F+4-4A$
Now I notice that I can simplify further if I set $G=F-A$ so that I can write the expression I want to minimise as $$16(F+1)^2+(3G+1)^2+4(2G-F+2)^2$$
If I fix $F$ and minimise with respect to $G$ I find I need $$6(3G+1)+8(2G-F+2)=0=34G-16F+22$$ so that $$G=frac {8F-11}{17}$$
and I continue by substituting back for $G$, and then work back from $G$ and $F$ to $A$ and $E$, having simplified the computation somewhat.
$endgroup$
I would do this differently, given you have a point on the curve.
First set $f(x)=A(x-4)(x-E)+4$ to give the point on the curve and to reduce the number of variables. We will put $AE=F$, since $E$ only appears as a product with $A$
Then
$f(0)=4F+4$,
$f(1)-3=-3A(1-E)+4-3=3F+1-3A$
$f(2)-5=-2A(2-E)+4-5=2F+4-4A$
Now I notice that I can simplify further if I set $G=F-A$ so that I can write the expression I want to minimise as $$16(F+1)^2+(3G+1)^2+4(2G-F+2)^2$$
If I fix $F$ and minimise with respect to $G$ I find I need $$6(3G+1)+8(2G-F+2)=0=34G-16F+22$$ so that $$G=frac {8F-11}{17}$$
and I continue by substituting back for $G$, and then work back from $G$ and $F$ to $A$ and $E$, having simplified the computation somewhat.
answered Jan 6 at 22:44
Mark BennetMark Bennet
80.8k981179
80.8k981179
$begingroup$
You wrote: "If I fix F and minimise with respect to G I find I need" I do not understand why that is a valid thing to do. Are you taking a partial derivative?
$endgroup$
– Bob
Jan 6 at 23:36
$begingroup$
I am thinking that you have only one equation and you need a second one. That is, you have two unknowns.
$endgroup$
– Bob
Jan 6 at 23:41
1
$begingroup$
@Bob The transformations I am making are reversible, so that a minimum value in respect of one set of parameters is also a minimum in respect of another. This is easy here because I am minimising a positive definite quadratic form. In terms of the equations, the point given on the curve gives one relation between the coefficients and reduces the number of necessary parameters to two. I have chosen to reduce and simplify before doing any other calculations. Your method keeps $A, B, C$
$endgroup$
– Mark Bennet
Jan 7 at 7:01
1
$begingroup$
@Bob Yes, I am taking partial derivative with respect to $G$. Since the form here is positive definite, it is easy to see that any local extremum is a minimum. Since it is linearly parametrised by two parameters ($A, F$ or $F, G$) it turns out that there is a single local minimum which is a global minimum. I also need to take the partial derivative with respect to $F$ to finish (or I can complete the square, which comes to the same thing). Since there is a single local minimum we can (with just a little care) locate it using the variables we choose rather than those we started with.
$endgroup$
– Mark Bennet
Jan 7 at 7:06
add a comment |
$begingroup$
You wrote: "If I fix F and minimise with respect to G I find I need" I do not understand why that is a valid thing to do. Are you taking a partial derivative?
$endgroup$
– Bob
Jan 6 at 23:36
$begingroup$
I am thinking that you have only one equation and you need a second one. That is, you have two unknowns.
$endgroup$
– Bob
Jan 6 at 23:41
1
$begingroup$
@Bob The transformations I am making are reversible, so that a minimum value in respect of one set of parameters is also a minimum in respect of another. This is easy here because I am minimising a positive definite quadratic form. In terms of the equations, the point given on the curve gives one relation between the coefficients and reduces the number of necessary parameters to two. I have chosen to reduce and simplify before doing any other calculations. Your method keeps $A, B, C$
$endgroup$
– Mark Bennet
Jan 7 at 7:01
1
$begingroup$
@Bob Yes, I am taking partial derivative with respect to $G$. Since the form here is positive definite, it is easy to see that any local extremum is a minimum. Since it is linearly parametrised by two parameters ($A, F$ or $F, G$) it turns out that there is a single local minimum which is a global minimum. I also need to take the partial derivative with respect to $F$ to finish (or I can complete the square, which comes to the same thing). Since there is a single local minimum we can (with just a little care) locate it using the variables we choose rather than those we started with.
$endgroup$
– Mark Bennet
Jan 7 at 7:06
$begingroup$
You wrote: "If I fix F and minimise with respect to G I find I need" I do not understand why that is a valid thing to do. Are you taking a partial derivative?
$endgroup$
– Bob
Jan 6 at 23:36
$begingroup$
You wrote: "If I fix F and minimise with respect to G I find I need" I do not understand why that is a valid thing to do. Are you taking a partial derivative?
$endgroup$
– Bob
Jan 6 at 23:36
$begingroup$
I am thinking that you have only one equation and you need a second one. That is, you have two unknowns.
$endgroup$
– Bob
Jan 6 at 23:41
$begingroup$
I am thinking that you have only one equation and you need a second one. That is, you have two unknowns.
$endgroup$
– Bob
Jan 6 at 23:41
1
1
$begingroup$
@Bob The transformations I am making are reversible, so that a minimum value in respect of one set of parameters is also a minimum in respect of another. This is easy here because I am minimising a positive definite quadratic form. In terms of the equations, the point given on the curve gives one relation between the coefficients and reduces the number of necessary parameters to two. I have chosen to reduce and simplify before doing any other calculations. Your method keeps $A, B, C$
$endgroup$
– Mark Bennet
Jan 7 at 7:01
$begingroup$
@Bob The transformations I am making are reversible, so that a minimum value in respect of one set of parameters is also a minimum in respect of another. This is easy here because I am minimising a positive definite quadratic form. In terms of the equations, the point given on the curve gives one relation between the coefficients and reduces the number of necessary parameters to two. I have chosen to reduce and simplify before doing any other calculations. Your method keeps $A, B, C$
$endgroup$
– Mark Bennet
Jan 7 at 7:01
1
1
$begingroup$
@Bob Yes, I am taking partial derivative with respect to $G$. Since the form here is positive definite, it is easy to see that any local extremum is a minimum. Since it is linearly parametrised by two parameters ($A, F$ or $F, G$) it turns out that there is a single local minimum which is a global minimum. I also need to take the partial derivative with respect to $F$ to finish (or I can complete the square, which comes to the same thing). Since there is a single local minimum we can (with just a little care) locate it using the variables we choose rather than those we started with.
$endgroup$
– Mark Bennet
Jan 7 at 7:06
$begingroup$
@Bob Yes, I am taking partial derivative with respect to $G$. Since the form here is positive definite, it is easy to see that any local extremum is a minimum. Since it is linearly parametrised by two parameters ($A, F$ or $F, G$) it turns out that there is a single local minimum which is a global minimum. I also need to take the partial derivative with respect to $F$ to finish (or I can complete the square, which comes to the same thing). Since there is a single local minimum we can (with just a little care) locate it using the variables we choose rather than those we started with.
$endgroup$
– Mark Bennet
Jan 7 at 7:06
add a comment |
Thanks for contributing an answer to Mathematics Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
Use MathJax to format equations. MathJax reference.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3064435%2ffinding-an-interpolating-polynomial-based-upon-four-points%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
1
$begingroup$
Error: sign of B is wrong, should be +.
$endgroup$
– herb steinberg
Jan 6 at 22:54
$begingroup$
@herbsteinberg I agree that I have a sign error for $B$. However, I still get a wrong value for $B$.
$endgroup$
– Bob
Jan 6 at 23:14
$begingroup$
Did you recalculate A and C, both of which use the value of B? B$=+frac{1714}{12936}$.
$endgroup$
– herb steinberg
Jan 7 at 2:48