src/data/missing-data.lisp

   1 ;;; -*- mode: lisp -*-
   2
   3 ;;; Time-stamp: <2009-09-17 17:08:47 tony>
   4 ;;; Creation:   <2009-03-12 17:14:56 tony>
   5 ;;; File:       missing-data.lisp
   6 ;;; Author:     AJ Rossini <blindglobe@gmail.com>
   7 ;;; Copyright:  (c)2009--, AJ Rossini.  BSD, LLGPL, or GPLv2, depending
   8 ;;;             on how it arrives.
   9 ;;; Purpose:    Template header file
  10
  11 ;;; What is this talk of 'release'? Klingons do not make software
  12 ;;; 'releases'.  Our software 'escapes', leaving a bloody trail of
  13 ;;; designers and quality assurance people in its wake.
  14
  15 ;;; This organization and structure is new to the 21st Century
  16 ;;; version..   Think, "21st Century Schizoid Man".
  17
  18
  19
  20 ;;; Missing data handling is critical - initial thought for this is to
  21 ;;; have a class which provides scalars which have type "missing" and
  22 ;;; perhaps the "supposed" statistical-typing, i.e. continuous ratio,
  23 ;;; ordinal, nominal, etc.   Also have metadata to describe where this
  24 ;;; came from.
  25 ;;;
  26 ;;; Then we can gensym a value per-dataset per-missing type.  This
  27 ;;; strategy (using gensyms to represent a particular
  28 ;;; missing-data-type instance could be useful if we do not have to
  29 ;;; maintain equality across CL execution runs.
  30 ;;;
  31 ;;; Different types of missing:
  32 ;;; * censored data
  33 ;;; * unobserved data
  34 ;;; * coarsened measurement data
  35 ;;;
  36 ;;; but they can be placed into a generalized framework (see the work
  37 ;;; of van der Laan, Robins, etc...).
  38 ;;;
  39