List of Routines By Functionality |  MISR Toolkit: Main Page

-Abstract

   MTK_LINEAR_REGRESSION_CALC Use linear regression to fit a set of observations (x,y) to the model: y(x) = a + b * x

-Copyright

   Copyright (2013), California Institute of Technology.
   U.S. Government sponsorship acknowledged.

-I/O

	Given: 
	num            number of elements
	x              x values
	y              y values
	y_sigma        uncertainty in y values
	
	the call:

	  status = MTK_LINEAR_REGRESSION_CALC( num, x, y, y_sigma, a, b, corr )

	returns: 
	status         0 on success; otherwise failure
	a              variable a
	b              variable b
	corr           correlation for linear regression
   
   
-Examples


	;;
	;; Set up input parameters
	;;
	num = 4	
	x = [1,5,7,8]
	y = [12,14,19,21]
	y_sigma = [0.1, 0.2, 0.2, 0.3]
	;;
	;; The call
	;;
	status = MTK_LINEAR_REGRESSION_CALC( num, x, y, y_sigma, a, b, corr )
	;;
	;; Output...
	;;
	print, '================================================='
	print, a, b, corr

	IDL outputs:
	=================================================
    10.696193       0.0000000       0.0000000       0.0000000
    1.0944363       0.0000000       0.0000000       0.0000000
    0.94802655       0.0000000       0.0000000       0.0000000
   
     

-Particulars
   None.

-Required Reading
For important details concerning this module's function, please refer to
the MTK routine MtkLinearRegressionCalc.c.


-Version
   -IDL-MTK Version 1.2.4