[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference hydra::amiga_v1

Title:AMIGA NOTES
Notice:Join us in the *NEW* conference - HYDRA::AMIGA_V2
Moderator:HYDRA::MOORE
Created:Sat Apr 26 1986
Last Modified:Wed Feb 05 1992
Last Successful Update:Fri Jun 06 1997
Number of topics:5378
Total number of notes:38326

2182.0. "PlotXY" by LEDS::ACCIARDI (Man Dies Laughing - Film at 11:00) Wed Feb 01 1989 03:36

    
    Yet another example of outstanding user-written Amiga software
    (YAEOOUWAS) is PlotXY.  Rather than gaphing a mathematical function,
    like many programs do, PlotXY takes your data file and applies
    some pretty useful analysis and curve fitting tool.  Rather than having me
    babble on about it, here's some excerpts from the included
    documentation...
    
    >Linear Regression - Linear regression allows you to analyze one
    >variable's relationship to another.  The method is to perform a
    >least-squares linear regression which is designed to minimize the sum
    >of the squares of the deviations of the actual data points from the
    >straight line of best fit. In practice, a plot of the data points is
    >made and then a line is constructed (by selecting the Linear Regression
    >Item) that uniformly divides the points.  The data color and line type
    >is chosen at this point.  The line is described by y = mx + b, where m
    >is the slope of the line and b is the intercept.  The correlation
    >coefficient, which is a measurement of how well the line fitted to the
    >data actually does approximate the data, is also calculated. These
    >parameters are displayed in a separate window. 
    
    >Polynomial Fit - Polynomial fit uses a higher order polynomial for a
    >reasonably good representation of a series of n data points (X,Y).  The
    >regression coefficients are also estimated.  The technique employs a
    >least-squares fit of the data by a polynomial of order m, where m =
    >1,2,...10 (maximum of 10) and examines the standard deviation s about
    >the regression line.  The linear regression of Y upon a single variable
    >X can be extended to the multiple regression: Y = a + b1X1 + b2x2 + ...
    >+ biXi + ... + bkXk, where X1,X2,...,Xk are k different variables. If,
    >as frequently occurs, X1 = X, X2 = X**2, X3 = X**3, etc., we have a
    >special case of multiple regression known as polynomial or curvilinear
    >regression. 
    
    >After the data points have been plotted, select the Polynomial Fit item
    >under the Analysis menu. The user is then prompted, via a string
    >gadget, to enter the order of the polynomial that will be used. The
    >color and line type are also selected. A curve is then plotted on the
    >graph.  A separate window is presented that displays: 1) the polynomial
    >order, 2)the standard deviation, 3) the coefficient a, and 4) each bk
    >regression coefficient. 
    
    The only weakness that I can see is that the data file can only
    contain one set of dependant and one set of independant variables,
    so you can't plot several columns of data that way that MaxiPlan
    or Analyze! can.  However, there's a simple way to display many curves
    on the screen at once.  Since the screen does not erase a curve
    until you force it to, you just feed it as many data tables as you
    need.  New curves are just painted to the screen over the old ones
    so that they're all visible.
    
    If you're into sexy 3D charts, forget it.  Simple 2D is all you
    get.  This program is no threat to Cricket Graph on the Mac, but
    what do you want for free?  There's actually quite a bit of flexibility
    in the display format, with cute little lables that you can paste
    around on the graph.  The author compiled the program to
    use Amy's IEEE math libs, so math chip freaks can finally grin.
                           
    The archive is in NORSE""::AMIGA:[UPLOAD] for now, in 128 byte format.
    
    Ed.
                                         
    
T.RTitleUserPersonal
Name
DateLines