Chapters: 1: Introduction 2: Simple example 3: Invocation 4: Finer Control 5: XY Plots 6: Contour Plots 7: Image Plots 8: Examples 9: Gri Commands 10: Programming 11: Environment 12: Emacs Mode 13: History 14: Installation 15: Gri Bugs 16: Test Suite 17: Gri in Press 18: Acknowledgments 19: License Indices: Concepts Commands Variables 
9.3.34: `

`regress {y vs x [linear]}{x vs y [linear]}' 
Perform linear regression of `y
' as a function of `x
' or `x
' as
a function of `y
'.
regress y vs x
'
Linear regression of y vs x. Several quantities are reported and also
saved into builtin variables. The intercept is defined as
`..coeff0..
', its 95 percent confidence limit is defined as
`..coeff0_sig..
'. Thus the confidence range is
`..coeff0....coeff0_sig..
' to `..coeff0..+..coeff0_sig..
'.
Similarly the slope and confidence limit are stored in `..coeff1..
'
and `..coeff1_sig..
'
The squared correlation coefficient is stored in `..R2..
'.
Historical note prior to version 2.1.15, a different meaning was
attached to `..coeff0_sig..
' and `..coeff1_sig..
'; they used
be defined as standard errors, without having been multiplied by the
appropriate studentt coefficient.
regress x vs y
'
Linear regression of x vs y; for notation see above.
regress y vs x linear
'
Linear regression of y vs x; for notation see above.
regress x vs y linear
'
Linear regression of x vs y; for notation see above.