Tail Probability Calculator Files

MMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
M                                                                     M
M       T a i l   P r o b a b i l i t y   C a l c u l a t o r         M
M       =====================================================         M
M                ReadMe relase notes fore version 2.0                 M
M                                                                     M
MMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM

    
Purpose:
========
This project contains a suite of Matlab functions that calculate the 
tail probability / CDF / PDF / quantile of linear combination of random 
variables in one of the following classes:
(1) symmetric random variables defined on the whole real axis (normal, 
    Student's t, uniform and triangular);
(2) random variables with support on the positive real axis (central and non-
    central chi-squared, standard and non-standard Lambda W x chi-squared
    distributions [added in version 2.0] and inverse gamma distributions 
    [temporarily disabled in version 2.0 due to numerical issues]).

For more information see
https://sourceforge.net/p/tailprobabilitycalculator/wiki/Home/


Copyright (c) 2014, Tomy Duby and Viktor Witkovsky
tomyduby@gmail.com and witkovsky@savba.sk


Installation
============
* Copy the downloaded Matlab files into a new folder on your computer.
* Make sure that this folder is on your Matlab path (use Matlab function 
  addpath to add this folder to your current path).


Revision History:
=================
Version 2.0, 31-oct-14
----------------------
Ticket #44: Introduce the log-Lambert W x chi-squared distributions. Adopting 
            the inversion algorithm for this distribution. 


Version 1.1, 5-Jun-14         
---------------------
Ticket #33: improved on the accuracy of pdf and cdf calculation for Student's t
            distribution. The results for pdf have now an absolute error < 1e-13
            and for cdf < 1e-9.
Ticket #38: Bug fix: TPC crashes when calculating the linear combination of
            two inverse gamma random variables.  


Version 1.0, 5-May-14
---------------------
Ticket #13: Introduced function TPC as the entry function to both the GUI and
            command line versions.
Ticket #17: Add license agreement to files.
Ticket #27: When the characteristic function contains only one element this 
            does not have to be a cell structure.
Ticket #28: Allow the 'Tail' optional argument accept 'cdf' and 'tail' values 
            in addition to "0" or "false" / "1" or "true".
Ticket #29: Validation of cdf, pdf and quantile calculation for non-central
            chi-squared distribution


Beta Version 0.5, 5-Apr-14
--------------------------
Ticket #20: Improving the accuracy of results of Student's t distribution
            for large values of arguments x.


Beta Version 0.4, 4-Apr-14
--------------------------
Mainly making sure that the software works and provides accurate results on a wide 
range of parameters for normal, Student's t, uniform, triangular, chi-squared and
inverse gamma distributions.
Ticket #18: Fixing bugs in validating calculations involving chi^2 distribution
Ticket #19: Fixing bugs in validating calculations involving inverse gamma distribution
Ticket #20: Fixing bugs in validating calculations involving Student's t distribution


Beta Version 0.3, 17-Feb-14
---------------------------
Ticket #14: Triangular distribution crashes in TPC_10. Fixed.
Ticket #15: TPC_10 fails to calculate pdf of five symmtric random variables.
Ticket #16: Allow to run TPC_20 with argument 'Version' or 'version' and 
  return the current software version.
This version runs with Example 4 of [Function TPC_20].
  

Beta Version 0.2, 8-Feb-14
--------------------------

This version fixes several bugs. Mainly ticket no. 8: Error when calculating 
the linear combination of random variables with support on positive real 
axis. I have thoroughly tested the program for one chi-squared or inverse
gamma distribution with either positive or negative coefficients of linear
combination and positive or negative values of x.