censored-continuous-class
(Package: mi) :
The "censored-continuous" Class, the "truncated-continuous" Class and Inherited Classes
The censored-continuous class and the truncated-continuous class are both virtual and both inherit from the continuous-class and each is the parent of four classes that differ depending on whether the lower and upper bounds are numeric vectors or functions. A censored observation is one whose exact value is not observed. A truncated observation is one whose exact value is not observed and which implies that values on some other variables are not observed for that unit of observation. An example of truncation might be that some taxation forms are not required when a person's income falls below a certain threshold. The methods for these classes are not working yet. Aside from these facts, the rest of the documentation here is primarily directed toward developeRs.
mi2stata
(Package: mi) :
Exports completed data in Stata (.dta) or comma-separated (.csv) format
This function exports completed data from an object of mi-class in which m completed data.frames are appended to the end of the raw data. Two additional variables are added which indicate the row number and distinguish the data.frames. The outputed file is either Stata (.dta) or comma-separated (.csv) format, and can be easily registered in Stata as multiply imputed data.
00mi-package
(Package: mi) :
Iterative Multiple Imputation from Conditional Distributions
The mi package performs multiple imputation for data with missing values. The algorithm iteratively draws imputed values from the conditional distribution for each variable given the observed and imputed values of the other variables in the data. The process approximates a Bayesian framework; multiple chains are run and convergence is assessed after a pre-specified number of iterations within each chain. The package allows customization of the conditional model and the treatment of missing values for each variable. In addition, the package provides graphics to visualize missing data patterns, to diagnose the models used to generate the imputations, and to assess convergence. Functions are included to run statistical models post-imputation with the appropriate degree of sampling uncertainty.
03change
(Package: mi) :
Make Changes to Discretionary Characteristics of Missing Variables
These methods change the family, imputation method, size, type, and so forth of a missing_variable. They are typically called immediately before calling mi because they affect how the conditional expectation of each missing_variable is modeled.
The mi function cannot be run in isolation. It is the most important step of a multi-step process to perform multiple imputation. The data must be specified as a missing_data.frame before mi is used to impute missing values for one or more missing_variables. An iterative algorithm is used where each missing_variable is modeled (using fit_model) as a function of all the other missing_variables and their missingness patterns. This documentation outlines the technical uses of the mi function. For a more general discussion of how to use mi for multiple imputation, see mi-package.
06pool
(Package: mi) :
Estimate a Model Pooling Over the Imputed Datasets
This function estimates a chosen model, taking into account the additional uncertainty that arises due to a finite number of imputations of the missing data.
The continuous class inherits from the missing_variable-class and is the parent of the following classes: semi-continuous, censored-continuous, truncated-continuous, and bounded-continuous. The distinctions among these subclasses are given on their respective help pages. Aside from these facts, the rest of the documentation here is primarily directed toward developers.
mi-internal
(Package: mi) :
Internal Functions and Methods
These functions are not intended to be called directly. In the case of methods, they documented elsewhere, either with the associated generic function or with the class of the object that the method is defined for.
categorical
(Package: mi) :
Class "categorical" and Inherited Classes
The categorical class is a virtual class that inherits from the missing_variable-class and is the parent of the unordered-categorical and ordered-categorical classes. The ordered-categorical class is the parent of both the binary and interval classes. Aside from these facts, the rest of the documentation here is primarily directed toward developers.
providing 
.