Description Usage Arguments Value Examples

The MICE Algorithm (Multiple Imputation by Chained Equations) is a method to impute missing data. This function uses this algorithm for imputing censored data, using inverse sampling to utilize the additional information given by the censored values.

1 2 3 4 |

`xmu_formula` |
formula. Formula for location parameter of xfamily. Dependent variable specifies the variable which is partially censored/missing and is to be imputed. |

`xsigma_formula` |
formula. Formula for scale parameter of gamlss family object. |

`xnu_formula` |
formula. Formula for skewness of gamlss family object. |

`xtau_formula` |
formula. Formula for kurtosis of gamlss family object. |

`xfamily` |
gamlss family object. Determines the family membership of the gamlss object. |

`data` |
data.frame. Input data frame containing a dummy variable as column, acting as an indicator; 1 if censored/missing, 0 if not. Note that in case of right (left) censoring, the censored variable contains the respective minimal (maximal) duration of the follow-up which is used for conditional imputation. In case of interval censored data, two columns are required; specifying the start and end durations of the interval in question. This implementation assumes, that the start duration is the observed time, in which no failure occured before the interval is entered. The exact point of the observed state change within the interval is unknown. For inverse sampling, the distribution is conditioned on the start-duration and cut at the end-duration to ensure the constraints. |

`indicator` |
character. Name of dummy column in data, which indicates the damaged observation. |

`censtype` |
character. The type of the damaged observation; 'missing', 'right', 'left' or 'interval'. |

`intervalstart` |
character. Name of the column of interval starting values. By convention, the starting duration in this column is assumed to be the time passed without failure, before entering the interval, in which the exact time of failure is unknown. |

`m` |
Number of imputations (How many rounds should the algorithm execute). Default is m = 5. |

`...` |
Additional arguments passed to all gamlss fits. |

Returns internal results of the algorithm.

1 2 3 4 5 6 7 8 9 | ```
# Simulating a dataset
missing = simulateData(n = 100, param.formula = list(mu = ~exp(x1) + x3,
sigma = ~sin(x2)), name = 'x1', subset = ~ x1 > 0.6, prob = 0.8,
damage = NA, family = 'NO', correlation = NULL)
# Imputing missing covariates
imputex(data = missing$defected, xmu_formula= x1~y+x3,
xsigma_formula = ~x2, xnu_formula = ~1, xtau_formula = ~1, xfamily =
NO(mu.link = 'identity'), indicator = "indicator", censtype= 'missing')
``` |

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.