From c532139fd6a79145828de5a924d34c856cdc7886 Mon Sep 17 00:00:00 2001 From: AJ Rossini Date: Thu, 17 Sep 2009 16:03:52 +0200 Subject: [PATCH] docs/ideas but no code for missing data. Signed-off-by: AJ Rossini --- src/data/missing-data.lisp | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) create mode 100644 src/data/missing-data.lisp diff --git a/src/data/missing-data.lisp b/src/data/missing-data.lisp new file mode 100644 index 0000000..f878e44 --- /dev/null +++ b/src/data/missing-data.lisp @@ -0,0 +1,36 @@ +;;; -*- mode: lisp -*- + +;;; Time-stamp: <2009-09-04 17:49:46 tony> +;;; Creation: <2009-03-12 17:14:56 tony> +;;; File: template.lisp +;;; Author: AJ Rossini +;;; Copyright: (c)2009--, AJ Rossini. BSD, LLGPL, or GPLv2, depending +;;; on how it arrives. +;;; Purpose: Template header file + +;;; What is this talk of 'release'? Klingons do not make software +;;; 'releases'. Our software 'escapes', leaving a bloody trail of +;;; designers and quality assurance people in its wake. + +;;; This organization and structure is new to the 21st Century +;;; version.. Think, "21st Century Schizoid Man". + + + +;;; Missing data handling is critical - initial thought for this is to +;;; have a class which provides scalars which have type "missing" and +;;; perhaps the "supposed" statistical-typing, i.e. continuous ratio, +;;; ordinal, nominal, etc. Also have metadata to describe where this +;;; came from. +;;; +;;; Then we can gensym a value per-dataset per-missing type. +;;; +;;; Different types of missing: +;;; * censored data +;;; * unobserved data +;;; * coarsened measurement data +;;; +;;; but they can be placed into a generalized framework (see the work +;;; of van der Laan, Robins, etc...). +;;; + -- 2.11.4.GIT