`FitCorCbm`

`DfCreateCorCbmDataset.Rd`

The paired dataset is generated using bivariate sampling; details are in referenced publication

```
DfCreateCorCbmDataset(
seed = 123,
K1 = 50,
K2 = 50,
desiredNumBins = 5,
muX = 1.5,
muY = 3,
alphaX = 0.4,
alphaY = 0.7,
rhoNor = 0.3,
rhoAbn2 = 0.8
)
```

- seed
The seed variable, default is 123; set to NULL for truly random seed

- K1
The number of non-diseased cases, default is 50

- K2
The number of diseased cases, default is 50

- desiredNumBins
The desired number of bins; default is 5

- muX
The CBM \(\mu\) parameter in condition X

- muY
The CBM \(\mu\) parameter in condition Y

- alphaX
The CBM \(\alpha\) parameter in condition X

- alphaY
The CBM

`alpha`parameter in condition Y- rhoNor
The correlation of non-diseased case z-samples

- rhoAbn2
The correlation of diseased case z-samples, when disease is visible in both conditions

The desired dataset suitable for testing `FitCorCbm`

.

The ROC data is bined to 5 bins in each condition.

Zhai X, Chakraborty DP (2017) A bivariate contaminated binormal model for robust fitting of proper ROC curves to a pair of correlated, possibly degenerate, ROC datasets. Medical Physics. 44(6):2207--2222.

```
## seed <- 1
## this gives unequal numbers of bins in X and Y conditions for 50/50 dataset
dataset <- DfCreateCorCbmDataset()
# \donttest{
## this takes very long time!! used to show asymptotic convergence of ML estimates
## dataset <- DfCreateCorCbmDataset(K1 = 5000, K2 = 5000)
# }
```