This is a collection of files that estimates an error bar and the autocorrelation time of
a time series that comes from a Markov chain.  

Author: Jonathan Goodman, goodman@cims.nyu.edu, http://www.cims.nyu.edu/faculty/goodman
Spring 2009.

Uses: 

1. On the command line, type: acor "file" to compute the mean, error bar, and autocorrelation
   time fo the time series in "file".  This file is assumed to have one floating point number
   per line, and nothing else.  For example, typing acor tSeries could give output:
   sample mean = 2.1345,  standard deviation = .034, autocorrelation time = 24.38

2. From a C++ program, call int acor( double *mean, double *sigma, double *tau, char* filename).
   This reads the file given in the last argument and returns the estimated information given.
   The return value is zero if it thinks it worked and nonzero otherwise.

3. From a C++ program, call int acor( double *mean, double *sigma, double *tau, double X[], int L).
   This works like the previous one, except that the numbers are in the array X passed, not read
   from a file.  L is the length of the array X.

Files:

acc.cpp  -- The C++ code that does the calculations.  It is the code for the acor procedure
            in use 3 above.
            AutoCorrelationCalculate

acrc.cpp -- A main program that reads a file specified in the command line sequence 
            then calls acor as given in use 3.  This is the front end code for use 1.
			AutoCorrelationReadCommandline

acre.cpp -- A procedure that reads a file specified in the calling sequence and calls
            acor as given in use 3.  This is the front end for use 2.
			
acor.h   -- A header file for the above

makefile -- The makefile for the above.  
            make acor -- makes the three 