Interactive statistical-distribution-analysis program utilizing numerical and graphical methods
The TERPED/P program is designed to facilitate the quantitative analysis of experimental data, determine the distribution function that best describes the data, and provide graphical representations of the data. This code differs from its predecessors, TEDPED and TERPED, in that a printer-plotter has been added for graphical output flexibility. The addition of the printer-plotter provides TERPED/P with a method of generating graphs that is not dependent on DISSPLA, Integrated Software Systems Corporation's confidential proprietary graphics package. This makes it possible to use TERPED/P on systems not equipped with DISSPLA. In addition, the printer plot is usually produced more rapidly than a high-resolution plot can be generated. Graphical and numerical tests are performed on the data in accordance with the user's assumption of normality or lognormality. Statistical analysis options include computation of the chi-squared statistic and its significance level and the Kolmogorov-Smirnov one-sample test confidence level for data sets of more than 80 points. Plots can be produced on a Calcomp paper plotter, a FR80 film plotter, or a graphics terminal using the high-resolution, DISSPLA-dependent plotter or on a character-type output device by the printer-plotter. The plots are of cumulative probability (abscissa) versus user-defined units (ordinate). The program was developed on a Digital Equipment Corporation (DEC) PDP-10 and consists of 1500 statements. The language used is FORTRAN-10, DEC's extended version of FORTRAN-IV.
- Research Organization:
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- DOE Contract Number:
- W-7405-ENG-26
- OSTI ID:
- 5255050
- Report Number(s):
- ORNL-5813; ON: DE82013341
- Resource Relation:
- Other Information: Portions of document are illegible
- Country of Publication:
- United States
- Language:
- English
Similar Records
Analysis/plot generation code with significance levels computed using Kolmogorov-Smirnov statistics valid for both large and small samples
User's manual for TEDPED, an interactive code for testing the hypothesis of log-normally or normally distributed data. [In FORTRAN IV for PDP-10]