% @(#)regresline.hlq	17.1.1.1 (ESO-IPG) 01/25/02 17:43:53
% @(#)regresline.hlq	8.1.1.1 (ESO-IPG) 8/31/94 12:16:50 
%++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
%.COPYRIGHT  (c)  1990 European Southern Observatory
%.IDENT      regresline.hlq
%.AUTHOR     JDP, IPG/ESO
%.KEYWORDS   MIDAS, help files, REGRESSION/LINEAR
%.PURPOSE    On-line help file for the command: REGRESSION/LINEAR
%.VERSION    1.0  17-OCT-1983 : Creation, JDP
%----------------------------------------------------------------
\se
SECTION./LINE
\es\co
REGRESSION/LINEAR					17-OCT-1983  JDP
\oc\su
REGRESSION/LINEAR table y x1,x2,...
	linear regression on table columns
\us\pu
Purpose:    
          Performs linear regression between  table columns.
\up\sy
Syntax:     
          REGRESSION/method table y x1,x2,... 
\ys\pa
          table = table name
\ap\pa
          y = column acting as dependent variable
\ap\pa
          x1,x2,... = columns acting as independent variables
\ap\out
Output:
          Output from the regression is listed on terminal and stored in 
          the following keywords:\\
          OUTPUTC   - character information\\
          OUTPUTI(1)- N,no.of data,\\
                 (2)- M,no.of ind.var.\\
                 (3)- col.no. of dep.var.\\
                 (i)- col.no. of ind.var.\\
                      i=4,...,M+3\\
          OUTPUTD(1)- Constant term\\
                 (i)- Coefficients\\
                      i=2,...,M+1\\
          OUTPUTR(1)- F-value for the analysis of variance\\
                 (3)- Standard error of the estimate\\
                 (i)- Standard error of coeffs.\\
                      i=4,...,M+3\\
		 if only one column given as independent
		 variable, OUTPUTR will be filled also with:     
		 (2)- Standard error of constant term\\
\tuo\sa
See also:
          SAVE/REGRESSION, COMPUTE/REGRESSION, REGRESSION/POLYNOMIAL
\as\no
Note:    
          Standard error is calculated with "1/(N-2)"!!!
\on\exs
Examples:
\ex
          REGRESSION/LINEAR data :Z :X,:Y
           Use the columns labeled :Z, :X, :Y of table `data.tbl' and fit the
           function :Z = a*:X + b*:Y + c to determine the values of a, b and c.
\xe\ex
          REGRESSION/LINEAR dito \#2 \#1
           Use the columns no. 2, no. 1 of table `dito.tbl' and fit the
           function \#2 = a*\#1 + b to determine the values of a and b.
\xe\sxe