NominalLogisticBiplot (version 0.2)

Generators: Generators (points) of the tesselation generated by a nominal variable.

Description

With the parameters resulting from fitting a nominal logistic model to the row scores for a given variable, the function calculates all the information necessary to plot the tessellation generated by the fit. The final user will not normally use this function.

Usage

Generators(beta)

Arguments

beta

Matrix with the estimated parameters for a given nominal variable. It has as many rows as the number of categories minus one and three columns (one for the constant and other two for the x-y coordinates on the plane).

Value

An object of class "voronoiprob". This has the components:

x

x-coordinates for the real points (Vertices of the tessellation).

y

y-coordinates for the real points (Vertices of the tessellation).

n1

vector with the first neighbours of the real points

n2

vector with the second neighbours of the real points

n3

vector with the third neighbours of the real points

dummy.x

x-coordinates for the dummy points

dummy.y

y-coordinates for the dummy points

ndummy

Number of dummies

IndReal

Matrix with the indices of each real point in the tessellation

Centers

Matrix with the points resulting from inverting the tessellation

hideCat

Vector to indicate if there are some hidden categories

equivRegiones

Matrix with the new re-numbered categories (when some are hidden)

References

Hern\'andez S\'anchez, J. C., & Vicente-Villard\'on, J. L. (2013). Logistic biplot for nominal data. arXiv preprint arXiv:1309.5486.

Gower, J. & Hand, D. (1996), Biplots, Monographs on statistics and applied probability 54. London: Chapman and Hall., 277 pp.

Evans, D. & Jones, S. (1987), Detecting voronoi (area of influence) polygons ,Mathematical Geology 19(6), 523--537.

Hartvigsen, D. (1992), Recognizing voronoi diagrams with linear programming, ORSA Journal on Computing 4, 369--374.

Schoenberg, F., Ferguson, T. & Li, C. (2003), Inverting dirichlet tesselations, Computer journal 46(1), 76--83.

Examples

  data(HairColor)
  data = data.matrix(HairColor)
  xEM = NominalLogBiplotEM(data, dim = 2,showResults = FALSE)
  nomreg = polylogist(data[,2],xEM$RowCoordinates[,1:2],penalization=0.1)
  tesselation = Generators(nomreg$beta)
  tesselation