Description Usage Arguments Value Author(s) References Examples

Set control parameters for the smoothing fit of stl and spatial smoothing

1 2 3 4 5 | ```
spacetime.control(vari = "resp", time = "date", n, stat_n, n.p = 12,
s.window, s.degree = 1, t.window = NULL, t.degree = 1, inner = 2,
outer = 1, mthbytime = 1, s.jump = 10, t.jump = 10, cell = 0.2,
degree, span, Edeg, surf = c("direct", "interpolate"),
family = c("symmetric", "gaussian"), siter = 2)
``` |

`vari` |
variable name in string of the response variable. The default is "resp" |

`time` |
variable name in string of time index of the whole time series. The default is "date". In the final results on HDFS, the index of time will be changed to this variable instead of year and month. |

`n` |
the number of total observations in the time series at each location. |

`stat_n` |
The number of stations. |

`n.p` |
the number of observations in each subseries. It should be 12 for monthly data for example. |

`s.window` |
either the character string |

`s.degree` |
degree of locally-fitted polynomial in seasonal extraction. Should be 0, 1, or 2. |

`t.window` |
the span (in lags) of the loess window for trend extraction, which should be odd. If |

`t.degree` |
degree of locally-fitted polynomial in trend extraction. Should be 0, 1, or 2. |

`inner` |
The iteration time for inner loop of stlplus for time dimension fitting |

`outer` |
The iteration time for outer loop of stlplus for time dimension fitting |

`mthbytime` |
The number of months will be grouped together in the by time division after |

`s.jump, t.jump` |
integers at least one to increase speed of the respective smoother. Linear interpolation happens between every '*.jump'th value. |

`cell` |
if interpolation is used this controls the accuracy of the approximation via the maximum number of points in a cell in the kd-tree. Cells with more than 'floor(n*span*cell)' points are subdivided. |

`degree` |
smoothing degree for the spatial loess smoothing. It can be 0, 1, or 2. |

`span` |
smoothing span for the spatial loess smoothing. |

`Edeg` |
the degree for the conditioanl parametric model including elevation. |

`surf` |
should the fitted surface be computed exactly or via interpolation from a kd tree. |

`family` |
if '"gaussian"' fitting is by least-squares, and if '"symmetric"' a re-descending M estimator is used with Tukey's biweight function. |

`siter` |
the number of iterations used for the spatial smoothing procedure if family is set to be '"symmetric"', which is for robust fitting. |

A list with space-time fitting parameters.

Xiaosu Tong

R. B. Cleveland, W. S. Cleveland, J. E. McRae, and I. Terpenning (1990) STL: A Seasonal-Trend Decomposition Procedure Based on Loess. *Journal of Official Statistics*, **6**, 3–73.

1 2 3 4 | ```
spacetime.control(
n = 576, stat_n = 7738, n.p = 12, s.window = 21, s.degree = 1, t.window = 241,
t.degree = 1, degree = 2, span = 0.015, Edeg = 2, surf = "interpolate"
)
``` |

Embedding an R snippet on your website

Add the following code to your website.

For more information on customizing the embed code, read Embedding Snippets.