Standard deviation calculation of a time series.

I wrote the following script to plot the Multiscale Entropy of a time series My_Signal in Matlab format:

>> load 'My_Signal.m'
>> x = val(1,:);
>> [entropy,scale]=msentropy(x);
>> subplot(2,1,1);
>> plot(x);
>> subplot(2,1,2);
>> plot(scale,entropy)

The function msetropy.m (from Physionet.org)is defined as follows:

function varargout=msentropy(varargin)
[y,scale,info]=msentropy(x,dn,dm,dr,N,N0,minM,maxM,maxScale,minR,maxR)

The input signal x is the only REQUIRED parameter and should be a Nx1 vector of doubles in which to calculate the multiscale entropy.

I would like minR to have the value 0.15*(Standard deviation of the original time series x ).

Questions:

1)How to implement the desired minR in Matlab ?

2)Since the input signal x should be a Nx1 vector of doubles, was it wise from me to use x=val(1,:) before making use of the function?

The input signals I am using are ECG data from Physionet.

Answer 1

[KSSV]

as x is your signal (NX1) vector

minR = 0.15*std(x) ;

Read about std

If your val is a matrix and if it has signal information in the first row, then

x = val(:,1)

should be used....