# Network Design

### Description

produces multi-resolution network.

### Usage

1 | ```
network.design(latlon, method = "Oh", type = "reduce", nlevel, x)
``` |

### Arguments

`latlon` |
grid points of observation sites in degree |

`method` |
network design method, "cover", "ModifyGottlemann", "Gottlemann" or "Oh" |

`type` |
specifies grid type, "regular" or "reduced" |

`nlevel` |
the number of observations in each resolution when using the method ‘"cover"’ |

`x` |
radius in degree |

### Details

This function partitions the grid points of observations into networks. Each network corresponds resolution level and level 1 is the most detailed level. Possible methods are

‘"cover"’ for utilizing ‘"cover.design"’ in the package ‘"field"’

‘"ModifyGottlemann"’ for modified G\"ottlemann's method

‘"Gottlemann"’ for G\"ottlemann's method

‘"Oh"’ for Oh's method.

For ‘"ModifyGottlemann"’, ‘"Gottlemann"’ and ‘"Oh"’, two types of design, ‘"regular"’ and ‘"reduced"’ are provided.

### Value

`netlab` |
vector of network labels indicating level of multi-resolution. |

### References

Oh, H-S. (1999) Spherical wavelets and their statistical analysis with applications to meteorological data. Ph.D. Thesis, Department of Statistics, Texas A\&M University, College Station.

### Examples

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 | ```
### Observations of year 1967
#data(temperature)
#names(temperature)
# Locations of 939 weather stations
#latlon <- temperature$latlon[temperature$year == 1967, ]
#netlab <- network.design(latlon=latlon, method="cover",
# nlevel=c(507, 244, 117, 49, 22))
#netlab <- network.design(latlon=latlon, method="ModifyGottlemann",
# type="regular", x=3)
#netlab <- network.design(latlon=latlon, method="Gottlemann",
# type="regular", x=2)
#netlab <- network.design(latlon=latlon, method="Oh",
# type="reduce", x=5)
``` |