Scribd
Upload
68 views6 pages

Curve Fitting With MATLAB

This document discusses using MATLAB to fit lines and polynomials to datasets. It explains how to use the linefit and polyfit functions to calculate the coefficients of the best fit line and polynomial through a sample x-y dataset. Higher order polynomials improve the R2 value compared to a linear fit. The polyval function is used to calculate the output of the fitted polynomial for given x values.

Uploaded by

emongoloide
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
68 views6 pages

Curve Fitting With MATLAB

This document discusses using MATLAB to fit lines and polynomials to datasets. It explains how to use the linefit and polyfit functions to calculate the coefficients of the best fit line and polynomial through a sample x-y dataset. Higher order polynomials improve the R2 value compared to a linear fit. The polyval function is used to calculate the output of the fitted polynomial for given x values.

Uploaded by

emongoloide
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

LinefittingwithMATLAB

MATLABhasabuiltinfunctionforgeneratingtheleastsquaresfitlinethrougha
dataset(x,y)
Thesyntaxis:
>>[c,R2]=linefit(x,y)
Uselinefitforthedataset:
x=[1020304050607080];
y=[257038055061012208301450];

>>[c,R2]=linefit(x,y)
c=
19.4702
234.2857
R2=
0.8805
Interpretationofthisoutcomeisthatthebestfitlinehastheequationy=
19.4702x234.287,thecoefficientofcorrelationR2valueforthisfit0.8805

Nextwewillexplorehigherorderpolynomialfitsforthesamedatasetand
hopefullyimprovetheR2value

IndealingwithpolynomialfunctionswithinMATLABweshouldusuallysubmitthe
coefficientsofthepolynomialsindescendingpowersofxasfollows:
f(x)=p1xn+p2xn1++pnx+pn+1
1|P a g e

Ifthereisamissingcoefficientinsertazeroinitsplace.
e.g.rootsofapolynomial,polynomialmanipulations,etc.

Manualfittingoflinesthroughdata

Computationofthefitparametersisaccomplishedbyimplementingthefollowing
treatmentoftermsinthedata

Example
Computemandbforthisdata

xi
1
2
3
4
5
15

yi
0.6
2.0
3.5
5.2
5.8
17.1

xi^2
1
4
9
16
25
55

yi^2
.36
4.0
12.25
27.04
33.64
77.29

xiyi
.6
4.0
10.5
20.8
29.0
64.9

2|P a g e

Applytheformulatocomputem
m=5(64.9)(15*17.1)/(5*5515^2)=1.36

b=(17.11.36*15)/(5)=0.66

Thusy=1.36x0.66

SimilarlycomputetheR2valuebycarefulsubstitution

3|P a g e

PolynomialfittingwithMATLAB
Herewewillexploretwocommonfunctionsthatusuallygohandinhand,polyfit
andpolyval
PolyfitisabuiltinMATLABfunctionforfittingpolynomialthroughdatasets.The
syntaxis:
>>polyfit(x,y,n)
Wherexandyareorderedpairsofthedatasetunderconsideration,andnisthe
orderofthepolynomialtobeused.Applyingthissyntaxdatasetgivenearlier:
>>a=polyfit(x,y,1)
a=
19.4702234.2857
Noticethatthetwocoefficientsechoedbythepolyfitcommandrepresenta
linearfunction
Theresultisidenticaltotheonegeneratedwiththelinefitfunction,thatisy=
19.4702x234.287,astraightlinewithaslopeof19.4072andayinterceptof
234.2857
Nowusepolyvaltocomputethenumericalvalueofthefitpolynomialusingthe
coefficientsaofthefitpolynomialfromthepolyfitfunctionwhenxtakesa
specificvaluee.g.45
>>polyval(a,45)
ans=
641.8750

4|P a g e

Nowtryusinghigherorderpolynomialstoimprovethefitquality
Hereisascriptusedforexperimentingwithpolynomialfitofthegivendataset.
Readthecommentsingreentocomprehendtheeffectofthecommandsused
%Polyfitdemonstrattion
x=[1020304050607080];
y=[257038055061012208301450];
%Generateasixthorderpolynomialfitthroughthedata
cp=polyfit(x,y,6)
%Defineadensearrayofdatafortherangeofxvalues
xfit=linspace(min(x),max(x));
%Defineadensearrayofdatafortherangeofyvalues
%usingthecpfitpolynomialcoefficientsatcorrespondingvaluesinthearray
%forthexfitvalues
yfit=polyval(cp,xfit);
%Generateaplotofthe
plot(x,y,'r.',xfit,yfit,'b')
grid
xlabel('x'),ylabel('y')
title('sixthorderpolynomialfitthroughthedataset')
legend('rawdata','fitdata')

Seeplotnextpage

5|P a g e

6|P a g e

You might also like