Construction and Selection of Smoothing Formulas

Abstract

Many types of phenomena are investigated by being observed at a large number of successive points or successive time intervals. All observations are subject to error; and it frequently happens that the errors of successive observations are such that the inferred velocities, accelerations, or the like, are wholly incredible as a result of these errors of observation. In cases of this sort some process of smoothing is a necessity; and even where no absurdity is obvious a smoothing process may be highly desirable. This means a process by which each observed value is altered slightly so as to bring it into reasonable relations with those that precede and follow it. In this there are obviously two conflicting aims, (a) to achieve reasonable smoothness, and (b) to alter the observed data by reasonably small amounts. The purpose of this report is to present a method by which these two conflicting aims are more effectively reconciled than by any previously used method.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 1953
Accession Number
AD0021009

Entities

People

  • L. S. Dederick

Organizations

  • Ballistic Research Laboratory

Tags

Communities of Interest

  • C4I
  • Energy and Power Technologies
  • Materials and Manufacturing Processes
  • Weapons Technologies

DTIC Thesaurus Topics

  • Air Force
  • Air Force Facilities
  • Binomials
  • California
  • Coefficients
  • Construction
  • Curve Fitting
  • Equations
  • Errors
  • Munitions
  • New Mexico
  • New York
  • Ordnance Laboratories
  • Polynomials
  • Reliability
  • Residuals
  • Sequences

Fields of Study

  • Mathematics

Readers

  • Approximation Theory.
  • Regression Analysis.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms