1
% @(#)regresline.hlq 19.1 (ESO-IPG) 02/25/03 14:03:39
2
% @(#)regresline.hlq 8.1.1.1 (ESO-IPG) 8/31/94 12:16:50
3
%++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
4
%.COPYRIGHT (c) 1990 European Southern Observatory
7
%.KEYWORDS MIDAS, help files, REGRESSION/LINEAR
8
%.PURPOSE On-line help file for the command: REGRESSION/LINEAR
9
%.VERSION 1.0 17-OCT-1983 : Creation, JDP
10
%----------------------------------------------------------------
14
REGRESSION/LINEAR 17-OCT-1983 JDP
16
REGRESSION/LINEAR table y x1,x2,...
17
linear regression on table columns
20
Performs linear regression between table columns.
23
REGRESSION/method table y x1,x2,...
27
y = column acting as dependent variable
29
x1,x2,... = columns acting as independent variables
32
Output from the regression is listed on terminal and stored in
33
the following keywords:\\
34
OUTPUTC - character information\\
35
OUTPUTI(1)- N,no.of data,\\
36
(2)- M,no.of ind.var.\\
37
(3)- col.no. of dep.var.\\
38
(i)- col.no. of ind.var.\\
40
OUTPUTD(1)- Constant term\\
43
OUTPUTR(1)- F-value for the analysis of variance\\
44
(3)- Standard error of the estimate\\
45
(i)- Standard error of coeffs.\\
47
if only one column given as independent
48
variable, OUTPUTR will be filled also with:
49
(2)- Standard error of constant term\\
52
SAVE/REGRESSION, COMPUTE/REGRESSION, REGRESSION/POLYNOMIAL
55
Standard error is calculated with "1/(N-2)"!!!
59
REGRESSION/LINEAR data :Z :X,:Y
60
Use the columns labeled :Z, :X, :Y of table `data.tbl' and fit the
61
function :Z = a*:X + b*:Y + c to determine the values of a, b and c.
63
REGRESSION/LINEAR dito \#2 \#1
64
Use the columns no. 2, no. 1 of table `dito.tbl' and fit the
65
function \#2 = a*\#1 + b to determine the values of a and b.