lstseq.kern {denpro} | R Documentation |

## Calculates a scale of kernel estimates

### Description

Calculates a scale of kernel estimates corresponding to a scale of smoothing
parmeters.

### Usage

lstseq.kern(dendat, hseq, N, lstree = NULL, level = NULL,
Q = NULL, kernel = "gauss", hw = NULL, algo = "leafsfirst", support = NULL)

### Arguments

`dendat` |
n*d matrix of real numbers; the data matrix |

`hseq` |
a vector of positive real numbers;
the sequence should be monotonic |

`N` |
vector of d positive integers; the dimension of the grid where the
kernel estimate will be evaluated; we evaluate the estimate on a regular
grid which contains the support of the kernel estimate |

`lstree ` |
if NULL, then level set trees are not calculated |

`level` |
NULL or a real number between 0 and 1;
if NULL, then shape trees are not calculated;
if number, then it is the level in percents of the maximum
of the level sets for which the shape trees are calculated |

`Q` |
positive integer; needed only in the DynaDecompose algorithm, see
parameter "algo"; the number of levels in the level set trees |

`kernel` |
"epane" or "gauss"; the kernel is either the
Bartlett-Epanechnikov product kernel or the standard Gaussian |

`hw` |
positive integer; parameter for time localized kernel estimation;
gives the smoothing parameter for the temporal smoothing |

`algo` |
"leafsfirst" or "dynadecompose" |

`support` |
2*d vector of reals gives the d intervals of a
rectangular support; c(low1,upp1,...,lowd,uppd) |

### Value

A list with components

`lstseq` |
a list of level set trees |

`pcfseq` |
a list of piecewise constant functions |

`stseq` |
a list of shape trees |

`hseq` |
a vector of smoothing parameters corresponding to the members
in the sequences |

### Author(s)

Jussi Klemela

### See Also

`scaletable`

### Examples

dendat<-sim.data(n=200,type="mulmod")
h1<-0.9
h2<-2.2
lkm<-5
hseq<-hgrid(h1,h2,lkm)
N<-c(16,16)
estiseq<-lstseq.kern(dendat,hseq,N)

[Package

*denpro* version 0.9.0

Index]