binseqtest-internal {binseqtest} R Documentation

## Internal functions

### Description

Internal functions, not to be called by user

### Usage

```
validAbparms(object)
validBound(object)
validBoundEst(object)
validBoundNBF(object)

abBindBothCalcK(object)
abtoBound(from)

pCalc(S,N,K,order,theta0=.5,alternative="two.sided",ponly=FALSE)
ciCalc(S,N,K,order,type="upper",alpha=0.025)

missNAbparms(ab,missN=NULL,...)

```

### Arguments

 `object` object, usually a boundary of some class `from` an object of class abparms `S` vector of number of successes `N` vector of number of trials `K` vector of number of ways to reach each bounary point `order` vector of ordering of boundary points `theta0` null value of probability of success each binary random variable `alternative` character, either 'two.sided', 'less', or 'greater' `ponly` logical, should only the specific p-value type given by alternative be calculated `type` character, type of one-sided confidence interval to calculate, either 'upper' or 'lower' `alpha` numeric, amount of error to allow on the one side of the confidence interval
 `ab` object of class 'abparms' `missN` numeric vector, the N values where assessments are missed `...` arguments passed to other functions, not used

### Details

The validXX functions check that the object is a valid member of the class XX. For example, validBound checks that a bound object is OK by sum the probability distribution using the N,S, and K values and checking that it is within computer error of 1. The validity checks are run automatically by the new() function as part of the S4 implementation.

The function `abBindBothCalcK` takes an abparms object and creates a bound object. It requires calculating K, which is the number of ways to reach each boundary point. It ignores the `binding` argument and assumes all boundaries are binding. The `abtoBound` function uses the `binding` argument to create either a `bound` object (for `binding`='both') or a `boundNBF` object otherwise. Users can use the `as` function to coerce an `abparms` object to a `bound` object.

The function `pCalc` takes a boundary and calculates p-values, and outputs a vector of p-values (ponly=TRUE) or list of 3 vectors (plower,pupper, pval).

The function `cCalc` takes a boundary and calculates one of the one sided confidence intervals as directed by the type argument (either 'upper' or 'lower').

The functions `analyzeBound` and `analyzeBoundNBF` take objects of the `bound` and `boundNBF` classes and create ones of the `boundEst` and `boundNBFEst` classes. This means basically that the confidence intervals and p-values are calculated that go with those bounds.

The functions `getAlternative` and `getTSalpha` get those parameters from the inputs.

The function `missNAbparms` modifies abparms objects to reflect missing assessments. This is the working function for the missN option in `modify`.

[Package binseqtest version 1.0.3 Index]