Home > chronux > spectral_analysis > hybrid > coherencycpb.m

coherencycpb

PURPOSE ^

Multi-taper coherency,cross-spectrum and individual spectra - continuous and binned point process data

SYNOPSIS ^

function [C,phi,S12,S1,S2,f,zerosp,confC,phistd,Cerr]=coherencycpb(data1,data2,params,fscorr)

DESCRIPTION ^

 Multi-taper coherency,cross-spectrum and individual spectra - continuous and binned point process data

 Usage:

 [C,phi,S12,S1,S2,f,zerosp,confC,phistd,Cerr]=coherencycpb(data1,data2,params,fscorr)
 Inputs: 
       data1 (continuous data in form samples x trials) -- required
       data2 (binned spike data in form samples x trials) -- required
       params: structure with fields tapers, pad, Fs, fpass, err, trialave
       - optional
           tapers : precalculated tapers from dpss or in the one of the following
                    forms: 
                   (1) A numeric vector [TW K] where TW is the
                       time-bandwidth product and K is the number of
                       tapers to be used (less than or equal to
                       2TW-1). 
                   (2) A numeric vector [W T p] where W is the
                       bandwidth, T is the duration of the data and p 
                       is an integer such that 2TW-p tapers are used. In
                       this form there is no default i.e. to specify
                       the bandwidth, you have to specify T and p as
                       well. Note that the units of W and T have to be
                       consistent: if W is in Hz, T must be in seconds
                       and vice versa. Note that these units must also
                       be consistent with the units of params.Fs: W can
                       be in Hz if and only if params.Fs is in Hz.
                       The default is to use form 1 with TW=3 and K=5

            pad            (padding factor for the FFT) - optional (can take values -1,0,1,2...). 
                    -1 corresponds to no padding, 0 corresponds to padding
                    to the next highest power of 2 etc.
                       e.g. For N = 500, if PAD = -1, we do not pad; if PAD = 0, we pad the FFT
                       to 512 points, if pad=1, we pad to 1024 points etc.
                       Defaults to 0.
           Fs   (sampling frequency) - optional. Default 1.
           fpass    (frequency band to be used in the calculation in the form
                                   [fmin fmax])- optional. 
                                   Default all frequencies between 0 and Fs/2
           err  (error calculation [1 p] - Theoretical error bars; [2 p] - Jackknife error bars
                                   [0 p] or 0 - no error bars) - optional. Default 0.
           trialave (average over trials when 1, don't average when 0) - optional. Default 0
       fscorr   (finite size corrections, 0 (don't use finite size corrections) or 1 
              (use finite size corrections) - optional (available only for spikes). Defaults 0.
 Outputs:
       C (magnitude of coherency - frequencies x trials if trialave=0; dimension frequencies if trialave=1)
       phi (phase of coherency - frequencies x trials if trialave=0; dimension frequencies if trialave=1)
       S12 (cross spectrum -  frequencies x trials if trialave=0; dimension frequencies if trialave=1)
       S1 (spectrum 1 - frequencies x trials if trialave=0; dimension frequencies if trialave=1)
       S2 (spectrum 2 - frequencies x trials if trialave=0; dimension frequencies if trialave=1)
       f (frequencies)     
       zerosp (1 for trials where no spikes were found, 0 otherwise)
       confC (confidence level for C at 1-p %) - only for err(1)>=1
       phistd - theoretical/jackknife (depending on err(1)=1/err(1)=2) standard deviation for phi 
                Note that phi + 2 phistd and phi - 2 phistd will give 95% confidence
                bands for phi - only for err(1)>=1 
       Cerr  (Jackknife error bars for C - use only for Jackknife -  err(1)=2)

CROSS-REFERENCE INFORMATION ^

This function calls: This function is called by:

SOURCE CODE ^

0001 function [C,phi,S12,S1,S2,f,zerosp,confC,phistd,Cerr]=coherencycpb(data1,data2,params,fscorr)
0002 % Multi-taper coherency,cross-spectrum and individual spectra - continuous and binned point process data
0003 %
0004 % Usage:
0005 %
0006 % [C,phi,S12,S1,S2,f,zerosp,confC,phistd,Cerr]=coherencycpb(data1,data2,params,fscorr)
0007 % Inputs:
0008 %       data1 (continuous data in form samples x trials) -- required
0009 %       data2 (binned spike data in form samples x trials) -- required
0010 %       params: structure with fields tapers, pad, Fs, fpass, err, trialave
0011 %       - optional
0012 %           tapers : precalculated tapers from dpss or in the one of the following
0013 %                    forms:
0014 %                   (1) A numeric vector [TW K] where TW is the
0015 %                       time-bandwidth product and K is the number of
0016 %                       tapers to be used (less than or equal to
0017 %                       2TW-1).
0018 %                   (2) A numeric vector [W T p] where W is the
0019 %                       bandwidth, T is the duration of the data and p
0020 %                       is an integer such that 2TW-p tapers are used. In
0021 %                       this form there is no default i.e. to specify
0022 %                       the bandwidth, you have to specify T and p as
0023 %                       well. Note that the units of W and T have to be
0024 %                       consistent: if W is in Hz, T must be in seconds
0025 %                       and vice versa. Note that these units must also
0026 %                       be consistent with the units of params.Fs: W can
0027 %                       be in Hz if and only if params.Fs is in Hz.
0028 %                       The default is to use form 1 with TW=3 and K=5
0029 %
0030 %            pad            (padding factor for the FFT) - optional (can take values -1,0,1,2...).
0031 %                    -1 corresponds to no padding, 0 corresponds to padding
0032 %                    to the next highest power of 2 etc.
0033 %                       e.g. For N = 500, if PAD = -1, we do not pad; if PAD = 0, we pad the FFT
0034 %                       to 512 points, if pad=1, we pad to 1024 points etc.
0035 %                       Defaults to 0.
0036 %           Fs   (sampling frequency) - optional. Default 1.
0037 %           fpass    (frequency band to be used in the calculation in the form
0038 %                                   [fmin fmax])- optional.
0039 %                                   Default all frequencies between 0 and Fs/2
0040 %           err  (error calculation [1 p] - Theoretical error bars; [2 p] - Jackknife error bars
0041 %                                   [0 p] or 0 - no error bars) - optional. Default 0.
0042 %           trialave (average over trials when 1, don't average when 0) - optional. Default 0
0043 %       fscorr   (finite size corrections, 0 (don't use finite size corrections) or 1
0044 %              (use finite size corrections) - optional (available only for spikes). Defaults 0.
0045 % Outputs:
0046 %       C (magnitude of coherency - frequencies x trials if trialave=0; dimension frequencies if trialave=1)
0047 %       phi (phase of coherency - frequencies x trials if trialave=0; dimension frequencies if trialave=1)
0048 %       S12 (cross spectrum -  frequencies x trials if trialave=0; dimension frequencies if trialave=1)
0049 %       S1 (spectrum 1 - frequencies x trials if trialave=0; dimension frequencies if trialave=1)
0050 %       S2 (spectrum 2 - frequencies x trials if trialave=0; dimension frequencies if trialave=1)
0051 %       f (frequencies)
0052 %       zerosp (1 for trials where no spikes were found, 0 otherwise)
0053 %       confC (confidence level for C at 1-p %) - only for err(1)>=1
0054 %       phistd - theoretical/jackknife (depending on err(1)=1/err(1)=2) standard deviation for phi
0055 %                Note that phi + 2 phistd and phi - 2 phistd will give 95% confidence
0056 %                bands for phi - only for err(1)>=1
0057 %       Cerr  (Jackknife error bars for C - use only for Jackknife -  err(1)=2)
0058 
0059 if nargin < 2; error('Need data1 and data2'); end
0060 if nargin < 3; params=[]; end;
0061 [tapers,pad,Fs,fpass,err,trialave,params]=getparams(params);
0062 clear params
0063 if nargin < 4 || isempty(fscorr); fscorr=0; end;
0064 
0065 if nargout > 7 && err(1)==0;
0066 %Cannot compute error bars if err(1)=0. Need to change params and run again.
0067     error('When errors are desired, err(1) has to be non-zero.');    
0068 end;
0069 if nargout > 9 && err(1)~=2; 
0070     error('Cerr computed only for Jackknife. Correct inputs and run again');
0071 end;
0072 [N,Ch]=check_consistency(data1,data2);
0073 zerosp=zeros(1,Ch); % intialize the zerosp variable
0074 
0075 nfft=max(2^(nextpow2(N)+pad),N);
0076 [f,findx]=getfgrid(Fs,nfft,fpass); 
0077 tapers=dpsschk(tapers,N,Fs); % check tapers
0078 J1=mtfftc(data1,tapers,nfft,Fs); % fourier transform of continuous data
0079 [J2,Msp2,Nsp2]=mtfftpb(data2,tapers,nfft); % fourier transform of point process data
0080 zerosp(Nsp2==0)=1; % set the trials where no spikes were found to have zerosp=1
0081 J1=J1(findx,:,:);
0082 J2=J2(findx,:,:);
0083 S12=squeeze(mean(conj(J1).*J2,2));
0084 S1=squeeze(mean(conj(J1).*J1,2));
0085 S2=squeeze(mean(conj(J2).*J2,2));
0086 if trialave; S12=squeeze(mean(S12,2)); S1=squeeze(mean(S1,2)); S2=squeeze(mean(S2,2)); end;
0087 C12=S12./sqrt(S1.*S2);
0088 C=abs(C12);
0089 phi=angle(C12);
0090 if nargout==10; 
0091   if fscorr==1; 
0092      [confC,phistd,Cerr]=coherr(C,J1,J2,err,trialave,[],Nsp2);
0093   else
0094      [confC,phistd,Cerr]=coherr(C,J1,J2,err,trialave);
0095   end;
0096 elseif nargout==9;
0097   if fscorr==1; 
0098      [confC,phistd]=coherr(C,J1,J2,err,trialave,[],Nsp2);
0099   else
0100      [confC,phistd]=coherr(C,J1,J2,err,trialave);
0101   end;
0102 end;
0103 clear Msp2

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