or.midp {epitools} R Documentation

## Odds ratio estimation and confidence intervals using mid-p method

### Description

Calculates odds ratio by median-unbiased estimation and exact confidence interval using the mid-p method (Rothman 1998).

### Usage

```or.midp(x, conf.level = 0.95, byrow = TRUE, interval = c(0, 1000))
```

### Arguments

 `x` input data can be 2x2 matrix or vector of length 4 `conf.level` confidence level (default is 0.95) `byrow` integer vectors are read in row-wise (default) `interval` interval for the `uniroot` that finds the odds ratio median-unbiased estimate and mid-p exact confidence interval for `oddsratio.midp`

### Details

Calculates odds ratio by median-unbiased estimation and exact confidence interval using the mid-p method (Rothman 1998, p. 251).

This function expects the following 2x2 table struture:

```              exposed   not exposed
disease  	 a1	    a0
no disease   b1	    b0
```

or a numeric vector of the form c(a1, a0, b1, b0).

This function is used by `oddsratio.midp`.

### Value

 `x` table that was used in analysis `data` same table as `x` but with marginal totals `estimate` median unbiased odds ratio `conf.level` confidence level used

### Author(s)

Tomas Aragon, aragon@berkeley.edu, http://www.phdata.science

### References

Kenneth J. Rothman and Sander Greenland (1998), Modern Epidemiology, Lippincott-Raven Publishers

`oddsratio`

### Examples

```##rothman p. 243
z1 <- matrix(c(12,2,7,9),2,2,byrow=TRUE)
z2 <- z1[2:1,2:1]
##jewell p. 79
z3 <- matrix(c(347,555,20,88),2,2,byrow=TRUE)
z4 <- z3[2:1,2:1]
or.midp(z1)
or.midp(z2)
or.midp(z3)
or.midp(z4)
```

[Package epitools version 0.5-10.1 Index]