Home > chronux > locfit > m > scb.m

# scb

## PURPOSE

Simultaneous Confidence Bands

## SYNOPSIS

function z=scb(x,y,varargin)

## DESCRIPTION

``` Simultaneous Confidence Bands

Example:
z = scb(E,NOx,'h',0.5);

result (z) is a matrix with four columns: evaluation points,
fitted values, lower confidence limit, upper confidence limit.
Most locfit arguments should work.```

## CROSS-REFERENCE INFORMATION

This function calls:
• kappa0 Compute the constants for `tube-formula' based simultaneous
• locfit Smoothing noisy data using Local Regression and Likelihood.
• predict Interpolate a fit produced by locfit().
This function is called by:

## SOURCE CODE

```0001 function z=scb(x,y,varargin)
0002
0003 % Simultaneous Confidence Bands
0004 %
0005 % Example:
0007 %   z = scb(E,NOx,'h',0.5);
0008 %
0009 % result (z) is a matrix with four columns: evaluation points,
0010 % fitted values, lower confidence limit, upper confidence limit.
0011 % Most locfit arguments should work.
0012
0013 fit = locfit(x,y,'ev','grid','mg',20,varargin{:});
0014 kap = kappa0(x,y,varargin{:});
0015 cb = predict(fit,'fitp','band','g','kappa',kap);
0016 z = [fit.fit_points.evaluation_points' cb{1} cb{3}];
0017
0018 return;```

Generated on Fri 28-Sep-2012 12:34:30 by m2html © 2005