Official websites use .gov
A .gov website belongs to an official government organization in the United States.
Secure .gov websites use HTTPS
A lock (
) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.
Summary
e-FITS (link currently only accessible to NIST staff) is a web-based tool used to generate graphs, tables, and random numbers for a large number of probability distributions. In addition, it can fit distributions to user-supplied data.
The methodolgy used to develop e-FITS is being extended to the e-Metrology project (link currently restricted to NIST staff). e-Metrology provides forms for common metrology problems encountered by NIST scientists and engineers.
RELATED PROGRAM / PROJECTS:
Description
DESCRIPTION:
e-FITS is a web-based tool, currently available to NIST staff, used to perform the following tasks for over 100 probability distributions.
- Generate graphs of probability functions (probability density, cumulative distribution, inverse cumulative distribution, hazard, cumulative hazard, survival, inverse survival).
- Generate tables for each of these probability functions.
- Generate random numbers from the specified distribution.
- Fit the distribution to user-supplied data. The fit analysis will include parameter estimation and diagnostic analysis of the fit.
e-Metrology is also a web-based tool currently available to NIST staff that can be used to perform the following tasks.
- Uncertainty following the ISO Guide to the Expression of Uncertainty in Measurement (GUM). Utilizes the R-based gummer routines written by Hung-Kung Liu, Will Guthrie and Antonio Possolo.
- Consensus means utilizing various methods. Consensus means are a key component of many SRM analyses.
- Interlaboratory analysis based on ASTM standard E-691 and proficiency testing based on ASTM standards E-2489A and E-2489B. Youden plots and bivariate normal tolerance region plots.
- A limit of detection analysis based on the proposed ASTM WK 19817. This implements a method developed by Andrew Rukhin, Stefan Leigh and Michael Verkouteren (of CSTL).
- Outlier detection for univariate normal data.
- Jim Filliben's 10-step analysis of full and fractional factorial designs.
- Linear and quadratic calibration and errors-in-variables regression.
- One and two factor analysis of variance with supporting graphics.