Download presentation
Presentation is loading. Please wait.
Published byMarvin Warner Modified over 9 years ago
1
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 1 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Bruce Mayer, PE Registered Electrical & Mechanical Engineer BMayer@ChabotCollege.edu Engr/Math/Physics 25 Chp6 MATLAB Regression
2
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 2 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Learning Goals cont
3
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 3 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Learning Goals cont Build Math Models for Physical Data using “n th ” Degree Polynomials Use MATLAB’s “Basic Fitting” Utility to find Math models for Plotted Data
4
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 4 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Scatter on Plots in XY-Plane A scatter plot usually shows how an EXPLANATORY, or independent, variable affects a RESPONSE, or Dependent Variable Sometimes the SHAPE of the scatter reveals a relationship Shown Below is a Conceptual Scatter plot that could Relate the RESPONSE to some EXCITITATION
5
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 5 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Linear Fit by Guessing The previous plot looks sort of Linear We could use a Ruler to draw a y = mx+b line thru the data But which Line is BETTER? and WHY?
6
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 6 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Least Squares Curve Fitting In a Previous Example polyfit(x,y,1) returned the Values of m & b How does PolyFit Make these Calcs? How Good is the fitted Line Compared to the Data? polyfit, as do most other curve fitters, uses the “Least Squares” Criterion
7
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 7 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Least Squares To make a Good Fit, MINIMIZE the |GUESS − data| distance by one of data Best Guess-y Best Guess-x
8
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 8 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Least Squares cont MATLAB polyfit Minimizes the VERTICAL distances; i.e.: Note that The Function J contains two Variables; m & b Recall from MTH1 that to MINIMIZE a Function set the 1 st (partial) Derivative(s) equal to Zero
9
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 9 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Least Squares cont To Minimize J, take The Two Partials yield Two LINEAR Eqns in m & b The two eqns can be solved EXACTLY for m & b the Book on pg 271 gives a good example Remember, at this point m & b are UNKNOWN xy 02 56 1011
10
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 10 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Least Squares cont In This Case Solving these Eqns for m & b yields m = 9/10 b = 11/6 This produces the best fit Line Taking ∂J/∂m = 0, and ∂J/∂b = 0 yields xy 02 56 1011
11
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 11 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Goodness of Fit The Distance from The Best-Fit Line to the Actual Data Point is called the RESIDUAL For the Vertical Distance the Residual is just δy If the Sum of the Residuals were ZERO, then the Line would Fit Perfectly Thus J, after finding m & b, is an Indication of the Goodness of Fit
12
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 12 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Goodness of Fit cont Now J is an indication of Fit, but we Might want to SCALE it relative to the MAGNITUDE of the Data For example consider –DataSet1 with x&y values in the MILLIONS –DataSet2 with x&y values in the single digits In this case we would expect J1 >> J2 To remove the affect of Absolute Magnitude, Scale J against the Data Set mean; e.g mean1 = 730 000 mean2 = 4.91
13
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 13 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Goodness of Fit cont The Mean-Scaling Quantity is the Actual-Data Relative to the Actual-Mean Finally the Scaled Fit-Metric, “r-squared’ As before the Squaring Ensures that all Terms in the sum are POSITIVE
14
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 14 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods r 2 = Coeff of Determination The r 2 Value is Also Called the COEFFICIENT OF DETERMINATION J Sum of Residual (errors) –May be Zero or Positive S Data-to-Mean Scaling Factor –Always Positive if >1 Data-Pt and data not “perfectly Horizontal” If J = 0, then there is NO Distance Between the calculated Line and Data Thus if J = 0, then r 2 = 1; so r 2 = 1 (or 100%) indicates a PERFECT FIT
15
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 15 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Meaning of r 2 The COEFFICIENT OF DETERMINATION Has This Meaning The coefficient of determination tells you what proportion of the variation between the data points is explained or accounted for by the best line fitted to the points. It indicates how close the points are to the line.
16
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 16 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Norm of Residuals MATLAB uses the Norm of Residuals as a Measure of Goodness of Fit The Norm of Residuals, N R, is simply the SqRt of J: Thus r 2 in Terms of N R : As a Measure of Goodness of Fit as the FIT Approaches Perfection J→0 so:
17
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 17 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods N R vs r 2 → Notice that r 2 is a RELATIVE Measure → it’s NORMALIZED to the WORST CASE Value of J which is S Thus r 2 can be expressed at PERCENTAGE withOUT Units N R is an ABSOLUTE measure that Technically Requires that it be stated with SAME UNITS as the dependent variable, y
18
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 18 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Data Scaling Data-Scaling is a SubTopic of DIMENSIONAL ANALYIS (DA) DA is Covered in 3 rd Yr ME/CE Courses –I Learned it in a Fluid Mechanics Course For Our Purposes we will cover only SCALING Sometimes we Collect Data with a SMALL Variation Relative to the Magnitude of the MEAN Leads to a SENSITIVE Analysis; e.g. This Data is Noisy During Analysis xy 89747313 89717309 89697310
19
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 19 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Data Scaling - Normalization The Significance of ANY Data Set Can be Improved by Normalizing Normalize Scale Data such that the Values run: 0 →1 0% → 100% Steps to Normalization 1.Find the MAX & MIN values in the Data Set; e.g., z max & z min 2.Calculate the Data Range, R D R D = (z max – z min ) 3.Calc the Individual Data Differences Relative to the MIN Δz k = z k - z min
20
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 20 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Data Scaling – Normailzation cont 4.Finally, Scale the Δz k relative to R D Ψ k = Δz k /R D 5.Scale the corresponding “y” values in the Same Manner to produce say, Φ k 6.Plot Φ k vs Ψ k on x & y scales that Run from 0→1 Example – Do Frogs Croak More on WARM Nites? Temperature (ºF) Croaks/Hr 88.620.0 71.616.0 93.319.8 84.318.4 80.617.1 75.215.5 69.714.7 82.017.1 69.415.4 83.316.2 78.615.0 82.617.2 80.616.0 83.517.0 76.314.1
21
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 21 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Normalization Example Normalize T → Θ CPH → Ω Now Compare Plots CPH vs T Ω vs Θ
22
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 22 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Plots Compared T-CPH Plot Ω-Θ Plot The Θ-Ω Plot Fully Utilizes Both Axes
23
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 23 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Basic Fitting Use MATLAB’s AutoMatic Fitting Utility to Find The Best Line for the the Frog Croaking Data SEE: Demo_Frog_Croak_BasicFit_1110.m
24
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 24 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods All Done for Today Croaking Frog
25
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 25 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Bruce Mayer, PE Licensed Electrical & Mechanical Engineer BMayer@ChabotCollege.edu Engr/Math/Physics 25 Appendix
26
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 26 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Linear Regression Tutorial Minimize Height Error δy See File ENGR- 25_Linear_Regressi on_Tutorial_1309.pp tx
27
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 27 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods
28
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 28 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Altitude of Right Triangle The Area of RIGHT Triangle The Area of an ARBITRARY Triangle By Pythagoras for Rt-Triangle
29
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 29 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Altitude of Right Triangle cont Solving for h Equating the A=½·Base·Hgt noting that
30
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 30 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Normalized Plot >> T = [69.4, 69.7, 71.6, 75.2, 76.3, 78.6, 80.6, 80.6, 82, 82.6, 83.3, 83.5, 84.3, 88.6, 93.3]; >> CPH = [15.4, 14.7, 16, 15.5, 14.1, 15, 17.1, 16, 17.1, 17.2, 16.2, 17, 18.4, 20, 19.8]; >> Tmax = max(T); >> Tmin = min(T); >> CPHmax = max(CPH); >> CPHmin = min(CPH); >> Rtemp = Tmax - Tmin; >> Rcroak = CPHmax - CPHmin; >> DelT = T - Tmin; >> DelCPH = CPH - CPHmin; >> Theta = DelT/Rtemp;DelCPH = CPH - CPHmin; >> Omega = DelCPH/Rcroak; >> plot(T, CPH,), grid >> plot(Theta,Omega), grid
31
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 31 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Start Basic Fitting Interface 1 FIRST → Plot the DATA
32
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 32 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Start Basic Fitting Interface 2 Expand Dialog Box Goodness of Fit; smaller is Better
33
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 33 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Start Basic Fitting Interface 3 Result Chk by polyfit >> p = polyfit(Theta,Ome ga,1) p = 0.8737 0.0429
34
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 34 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Caveat
35
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 35 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Greek Letters in Plots
36
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 36 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods Plot “Discoverables”
37
BMayer@ChabotCollege.edu ENGR-25_Plot_Model-4.ppt 37 Bruce Mayer, PE Engineering/Math/Physics 25: Computational Methods % "Discoverable" Functions Displayed % Bruce Mayer, PE ENGR25 15Jul09 % x = linspace(-5, 5); ye = exp(x); ypp = x.^2; ypm = x.^(-2); % plot all 3 on a single graphe plot(x,ye, x,ypp, x,ypm),grid,legend('ye', 'ypp', 'ypm') disp('Showing MultiGraph Plot - Hit ANY KEY to continue') pause % % PLot Side-by-Side subplot(1,3,1) plot(x,ye), grid subplot(1,3,2) plot(x,ypp), grid subplot(1,3,3) plot(x,ypm), grid
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.