kosta
Member
Joined: 09 Aug 2015
Posts: 161
Location: Salonic
|
 Posted: 02/12/2016 22:13 Post subject: Julia - NAG Integration |
 |
|
Julia - NAG Integration by Reid Atcheson and Andy Greenwell
About Julia Computing
Julia Computing was founded in 2015 by the co-authors of the Julia programming language to help private businesses, government agencies and others develop and implement Julia-based solutions to their big data and analytic s problems.
Julia is an open-source language for high-performance technical computing created by some of the best minds in mathematical and statistical computing.
Julia has quickly become the preferred programming language for data and analytic. Julia combines the functionality of quantitative environments such as MATLAB®, R, SPSS®, Stata®, SAS® and Python® with the speed of production programming languages like Java® and C++ to solve big data and analytics problems. It delivers dramatic improvements in simplicity, speed, capacity and productivity for data scientists, algorithmic traders, quants, scientists and engineers who need to solve massive computation problems quickly and accurately.
Julia offers an unbeatable combination of simplicity and speed that is thousands of times faster than other mathematical, scientific and statistical computing languages.
Introduction to Correlation Matrices
Correlations provide helpful information when working with possibly mutually related data sources, such as the value of stocks, options, precious metals or others. In the case of three or more data sources this produces a matrix known as a "correlation matrix" which has a special structure, shown in Julia code below:
| Code: |
#The input Y holds m observations of n random variables
C = cor(Y)
#Output C contains pairwise correlations of Y's columns
|
Correlation (or, to give its full name, "Pearson's product-moment correlation coefficient"), is a measure of the linear relationship between two random variables. One of the features that makes it particularly useful is that it is invariant to location-scale transformations of the variables, giving values on the interval [-1,1]: 1 being deplin positive linear relationship, 0 being no linear relationship (though, it should be emphasised, not necessarily independent), and -1 being a perfect negative relationship.
A correlation matrix is then the set of pairwise correlations between numeroase random variables, where the (i,j)th element is the correlation between the ith and the jth variable. Mathematically, a correlation matrix must satisfy the following properties:
it must be symmetric,
the diagonal elements must be 1, and
it must be positive semidefinite, which is equivalent to all eigenvalues being non-negative.
read more and delight on
http://www.nag.com/blog_files/nagjuliablog/index.html
References Trademark Usage
| Code: | Java® is a registered trademark of Oracle Corporation
MATLAB® is a registered trademark of The MathWorks, Inc.
Python® is a registered trademark of the Python Software Foundation
SPSS® is a registered trademark of IBM
Stata® is a registered trademark of StataCorp LP
SAS® is a registered trademark of SAS Institute Inc. |
Numerical Algorithms Group
Solving numerical, technical and high performance computing problems every day, all over the world.
http://www.nag.com/
http://www.nag.com/content/nag-library |
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
