Table of Contents

Name

linreg - six linear regressions

Synopsis

linreg [parameter=value]

Description

linreg computes the slope (a) and intercept coefficients (b) for for OLS(Y/X) (1), OLS(X/Y) (2), OLS bisector(3) , orthogonal(4) , reduced major axis (5) and mean OLS (6) methods. The dispersions are returned in the arrays siga and sigb.

The dispersions for OLS are the general form provided by T. Isobe, E. D. Feigelson, M. G. Akritas and G. J. Babu, Ap.J. 364, 104. (1990)

mean OLS is 0.5*[OLS(X/Y) + OLS(Y/X)]

Parameters

The following parameters are recognized in any order if the keyword is also given:
in=
Input ascii table
xcol=
Column for X coordinates
ycol=
Column for Y coordinates
mode=
Output mode (0=all, 1..6=various methods
maxline=
Maximum size of columns

See Also

linreg(3NEMO) , fitexy(3NEMO) , gaussfit(1NEMO) , tablsqfit(1NEMO)
T. Isobe,  E. D. Feigelson, M. G. Akritas and G. J. Babu,  (1990)
Ap.J. 364, 104.
Pogge and Peterson, (1992)
A.J. 103, 1084 - Appendix
Akritas, Michael G.; Bershady, Matthew A.  (1996)
Ap.J. 470,706.
http://www.astro.psu.edu/users/edf/research/stat.html (Astrostatistics)
Reilly, Reilly & Keeler 1993, Applied Statistics, vol. 42
Britt & Luecke 1973, Technometrics, vol. 15
Jeffreys 1980, ApJ, 85, 177
https://arxiv.org/abs/1704.05466

Author

Peter Teuben

Update History


23-May-96    Created        PJT


Table of Contents