Description Usage Arguments Details Value Author(s) Examples

View source: R/measure_compare.R

This function implements the methodology reported in the paper entitled "Effective plots to assess bias and precision in method comparison studies" published in Statistical Methods in Medical Research (P. Taffe, 2015) (to appear).

1 | ```
measure_compare(data, new = "y1", Ref = "y2", ID = "id")
``` |

`data` |
a dataframe containing the identification number of the subject (id), the measurement values from the new measurement method (y1) and those from the reference methods). |

`new` |
specify the variable name or location of the new measurement method |

`Ref` |
specify the variable name or location of the reference standard |

`ID` |
specify the variable name for location of the subject identification number |

This functions implements the new estimation procedure to assess the agreement between the two measurement methods, as well as Bland & Altman's limits of agreement extended to the setting of possibly heteroscedastic measurement errors.

The function return a list with the following iterms:

Bias: differential and proportional bias for new method and the associated 95 percent confidence intervals

Models: list of models fitted in estimation procedure

Ref: a data frame containing the various variables used to compute the bias and precision plots, as well the smooth standard errors estimates of the reference standard

New: a data frame containing the various variables used to compute the bias and precision plots, as well the smooth standard errors estimates of the new measurement method

Mingkai Peng

1 2 3 4 | ```
### Load the data
data(data1)
### Analysis
measure_model <- measure_compare(data1)
``` |

```
Loading required package: nlme
```

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.