Data Analysis & Simulation

Since fitting probability distributions to large data sets can be a time-consuming task, we are currently researching the possibility of using multi-core processors to make EasyFit work faster. During the past several years, major processor manufacturers have been promoting the multi-core technology on the desktop processors market. Multiple cores in a single chip allow for better performance/price ratio on a range of tasks, however, existing software needs to be updated accordingly to take full advantage of this type of hardware.

We have modified the original distribution fitting algorithm to utilize all cores available on a system, and used it to fit distributions to a simulated set of 200,000 data points. In a series of tests on an Intel dual-core processor, the new algorithm executed almost twice as fast, yielding up to 90% performance increase, compared to the version currently used in EasyFit. These are very good results, and we will definitely be including this feature into the next release of EasyFit.

On a related note, last week we were contacted by a customer regarding our upcoming Simulation & Probabilistic Analysis SDK. They need to analyze large volumes of data, and from their description of the problem we estimated that the typical analysis would take up to 20 hours on a modern PC. With the new distribution fitting algorithm, it can take less than 12 hours on a dual-core CPU, or even less on quad-core processors popular in the server space. In a decision making environment where several hours can mean the difference between profit and loss, this is a very important improvement.

Lately we have received a couple messages from customers asking if it’s possible to use the Monte Carlo simulation and distribution fitting features of EasyFit in their own software applications. The short answer is yes, but these features are limited to calculating some distribution functions in Excel VBA. There’s currently no way to run simulations, fit distributions to data, perform goodness of fit tests, or use distribution functions from C#, C++, VB.NET, and other programming languages.

To fill the gap, we are considering to create a Simulation & Probabilistic Analysis Software Development Kit (SPA SDK) for software developers who need to deal with risk & uncertainty in their applications, but don’t have time or expertise to design and implement the required features on their own. We already have in place the tried and true technology that’s a basis for our distribution fitting products EasyFit and EasyFitXL, so creating an SDK would be possible in a short period of time.

Since we have had only a few requests for an SDK, we would like to know whether you would be interested in such kind of product. Below is our vision for the SDK – you are welcome to express any thoughts or specific requirements you might have. Please feel free to contact us and we will take your input seriously.

What is a Simulation & Probabilistic Analysis Software Development Kit (SPA SDK) ? Read the rest of this entry »

EasyFit ships with a comprehensive help file providing detailed information on all aspects of fitting distributions to data and interpreting the analysis results. For instance, it includes the description of supported distributions, goodness of fit tests, and output graphs.

If you are considering to try EasyFit but not sure if it has a particular feature you need, you can refer to the EasyFit help online which we have made available on our website for your convenience. Of course, you are still welcome to contact us for any questions regarding EasyFit or fitting distributions in general.

From now on, our distribution fitting software EasyFit can be ordered on a CD, in addition to the download version. The motivation for adding this feature was that in some countries it is required for accounting purposes. Most times you won’t need a CD since EasyFit is available for immediate download from our website, but if you do need it, you might find this feature useful.

The CD will be produced exclusively for you, and will include the EasyFit installation package along with your personal license key and registration details in a separate file.

The CD is priced at $14.95, including shipping. To receive a CD, check the option “Order this product on a CD” and make sure you specify a valid postal address when filling in the order form. Once your order is processed, you will immediately receive your license key via email (so you can start using EasyFit right away), and the CD will be shipped to you in a few days.

We are glad to announce that customers from Italy can now purchase EasyFit, our distribution fitting software product, through SxST, the Milan-based company specializing in software solutions for science and technology.

SxST provides a wide range of scientific software products from recognized vendors to Italian businesses, government organizations and academic institutions. According to the agreement, SxST will be offering the Italian version of EasyFit and the first level support in Italian language to their clients, ensuring the best customer experience at the same cost as for the rest of EU countries.

It is really fascinating for us to see EasyFit standing in line with statistical software packages by Systat Software, Minitab, GraphPad, and StataCorp. Our partnership with SxST can be thought of as a milestone indicating that EasyFit has reached the quality level of brand name products offered by these long-established companies.

In EasyFit 3.0 - back in 2006 - we introduced StatAssist, the built-in distribution viewer tool that closely integrates with the distribution fitting features of EasyFit. Since then, StatAssist has proven to be quite a useful feature, so we decided to include it into EasyFitXL, our distribution fitting add-in for Excel.

StatAssist displays graphs and other useful properties of all the probability distributions available in EasyFit. Even though it has initially been designed as an essential part of EasyFit, StatAssist can be used as a separate application – for example, to take a quick look at a distribution curve, or to calculate the distribution statistics… read the full article

Excel has become the de facto standard application for data analysis and presentation across a variety of industries, so if you deal with random data of any kind, chances are your data is stored in Excel workbooks. However, analyzing probability data in Excel can be tricky as the standard Excel package includes no facilities for fitting probability distributions to data. That is when EasyFitXL, the distribution fitting add-in for Excel, comes in handy… read the full article

Since 1991, the European Union has been promoting the use of numerical models to assess the environmental fate and risk of pesticides. Recently a group of scientists from the Catholic University and the Marche Polytechnic University (Italy) in association with Informatica Ambientale, the Milan-based research and computer science company, developed a tool that integrates one of the pesticide fate models with GIS software. Several distribution fitting software products were tested to introduce distribution functions in the risk assessment study, and EasyFit was selected as the most appropriate tool for analyzing annual mean pesticide concentration and determining the most suitable distribution… read the full case study

We are pleased to announce that a new version of EasyFit is now available for download.

EasyFit and EasyFitXL (distribution fitting add-in for Excel) now include support for 15 new probability distributions used in economics, finance, actuarial science, hydrology, communications theory, mechanical engineering, and other fields. To learn more about new features in this release, click here.

Upgrading From Previous Versions

If you purchased EasyFit on 23 October 2007 or later, you are entitled to a free upgrade, otherwise you qualify for a discounted upgrade. In either case, please contact us for upgrade information.

It’s been a while since we updated our blog - that’s because we have been working on a new version of EasyFit planned to be released later in October 2008. Below is a quick summary of new features in EasyFit.

(Note: The price for the new version is likely to increase, but since all licenses come with one year of free updates, you can still order EasyFit at the current price this month and upgrade to the upcoming new version for free.)

New Probability Distributions

EasyFit will include 15 new distributions:

The Burr (Singh-Maddala), Dagum (Inverse Burr), Pareto Type II (Lomax), and Log-Gamma distributions are mainly used in economics and actuarial science. We have had numerous requests for these models from some of our customers - actuaries and finance industry professionals.

The Generalized Gamma distribution is an advanced 4-parameter model including the Gamma, Weibull, and Half-Normal distributions as special cases.

The Pearson Type V (Inverse Gamma) and Pearson Type VI (Beta Prime) distributions are a part of the Pearson family of distributions often used in financial applications.

The Levy distribution is one of the few distributions that are stable and that have probability density functions that are analytically expressible, the others being the Normal distribution and the Cauchy distribution.

The Log-Pearson III (often referred to as “LP3″) distribution is the standard model recommended by the U.S. Water Resources Council for use by all U.S. Government agencies for flood frequency analysis.

The Nakagami (Nakagami-m) and Rice (Nakagami-n) distributions are used in communications theory to model signal fading under specific conditions.

The Kumaraswamy distribution is used in electrical, civil, mechanical, and financial engineering applications dealing with variables that are lower and upper bounded.

Some other new models include the Hyperbolic Secant, Error (Exponential Power), and Reciprocal distributions.

New Excel Functions

In EasyFit 4.0, we added the ability to analyze probability data in Excel, which has proven to be quite a useful feature. EasyFit 5.0 (Professional Edition) extends this functionality by introducing several new functions making it even easier to use distributions in your worksheet models.

The DistFit function allows to fit a specified distribution to your data and display the resulting model right in a worksheet. This function can be useful if you already know the underlying distribution (e.g. the Normal model) and only need to estimate the distribution parameters. The return value of this function can be supplied to several other new functions such as DistPdf, DistCdf, DistRand etc. For example, specifying

=DistCdf(DistFit(”Normal”; B1:B100); 2)

in a worksheet will fit the Normal distribution to the data in cells B1:B100, and then evaluate the CDF of the resulting Normal distribution at x=2.