Matlab 4 parameters&nb…

Matlab 4参数的逻辑回归分析

Four Parameters logistic regression

    The Four Parameters Logistic Regression or 4PL nonlinear regression model is commonly used for curve-fitting analysis in bioassays or immunoassays such as ELISAs or dose-response curves. 

    It is characterized by it classic or sigmoidal shape that fits the bottom and top plateaus of the curve, the EC50, and the slope factor (Hill slope). This curve is symmetrical around its inflection point. 

 The 4PL equation is:

  F(x) = D+(A-D)/(1+(x/C)^B)

  where:

  A = Minimum asymptote. In a bioassay where you have a standard curve,

  this can be thought of as the response value at 0 standard concentration.

 

  B = Hill's slope. The Hill's slope refers to the steepness of the curve.

  It could either be positive or negative.

 

  C = Inflection point. The inflection point is defined as the point on the

  curve where the curvature changes direction or signs. C is the

  concentration of analyte where y=(D-A)/2.

 

  D = Maximum asymptote. In an bioassay where you have a standard curve,

  this can be thought of as the response value for infinite standard

  concentration. 

  

 

  Syntax: [cf G]=L4P(x,y,st,L,U)

  

  Inputs: 

            X and Y (mandatory) - data points.

            X is a Nx1 column vector and Y must have the same rows number

            of X. If Y is a NxM matrix (N points and M replicates), L5P

            will generate a  column vector computing means for each row.

            The standard deviations of the rows will be used as weights of

            regression.

 

            st = starting points. This is a 1x4 vector of starting points

            that have to be used to start the process of not linear

            fitting. If this vector is not provided, L4P will set the

            starting points on the basis of x and y data.

 

            L = Lower bounds of parameters. This is a 1x4 vector of lower

            bounds of the 4 parameters. If this vector is not provided, L4P

            will set it on the basis of x and y data.

 

            U = Upper bounds of parameters. This is a 1x4 vector of upper

            bounds of the 4 parameters. If this vector is not provided, L4P

            will set it on the basis of x and y data.

 

  Outputs:

            cf = the FIT object

            G = goodness-of-fit measures, for the given inputs, in the

            structure G. G includes the fields: 

            -- SSE         sum of squares due to error

            -- R2          coefficient of determination or R^2

            -- adjustedR2  degree of freedom adjusted R^2

            -- stdError    fit standard error or root mean square error

Example:

 

  x=[0 4.5 10.6 19.7 40 84 210]; y=[0.0089 0.0419 0.0873 0.2599 0.7074 1.528 2.7739];

 

  Calling on MatLab the function: [cf G]=L4P(x,y)

  

            Answer is:

  

  cf = 

  

       General model:

       cf(x) = D+(A-D)/(1+(x/C)^B)

       Coefficients (with 95% confidence bounds):

         A =    0.001002  (-0.04594, 0.04794)

         B =       1.515  (1.293, 1.738)

         C =         108  (86.58, 129.4)

         D =       3.784  (3.302, 4.266)

  

  G = 

  

             sse: 0.0012

         rsquare: 0.9998

             dfe: 3

      adjrsquare: 0.9996

            rmse: 0.0200

 

  hold on; plot(x,y,'ro'); plot(cf,'r'); hold off

  this will plot the curve.

 

相关问题：L4P相关知识，matlab-code