The system offers connectivity to hundreds of dealers and trading systems and most of the major prime brokers. It includes a highly configurable real-time P&L and advanced functions such as profit attribution and risk reporting. It can be provided as a plug-and-play appliance or hosted on our servers.

Through the use of new web-based technologies the software achieves unparalleled performance, functionality and value for money.

For more information please call +1 416 644 1861 or request a demo at http://www.sgggpsc.com

Python is a powerful programming language, but occasionally we need to solve problems that can more easily be solved in other, more specialised, programming languages.

For statistical analysis and data visualisation, the programming language R, is an excellent choice. But where R is a very strong language for mathematical (numerical) computation and statistics, it is rather cumbersome to perform typical scripting tasks with it.

Ideally, we would want to use Python for the scripting tasks and R for the statistical tasks. And for once, we can actually get the best of both worlds by combining the two languages.

Integrating Python and R is particularly simple due to the rpy module, that essentially embeds R as a module in Python.

Python in Finance

scikits.statsmodels

Statsmodels is a Python package that provides a complement to scipy for statistical computations including descriptive statistics and estimation of statistical models.

Main Features

Where to get it

Development branches will be on Github. This is where to go to get the most up to date code in the trunk branch. Experimental code is hosted here in branches and in developer forks. This code is merged to master often. We try to make sure that the master branch is always stable.

https://www.github.com/statsmodels/statsmodels

Source download of stable tags will be on SourceForge.

https://sourceforge.net/projects/statsmodels/

scikits.timeseries

The scikits.timeseries module provides classes and functions for manipulating, reporting, and plotting time series of various frequencies. The focus is on convenient data access and manipulation while leveraging the existing mathematical functionality in Numpy and SciPy.

la (larry: labeled array object)

The main class of the la package is a labeled array, larry. A larry consists of data and labels. The data is stored as a NumPy array and the labels as a list of lists (one list per dimension).

A larry can have any number of dimensions except zero.

larry has built-in methods such as ranking, merge, shuffle, move_sum, zscore, demean, lag as well as typical Numpy methods like sum, max, std, sign, clip. NaNs are treated as missing data.

Alignment by label is automatic when you add (or subtract, multiply, divide) two larrys. You can also specify the join method (inner, outer, left, right) for binary operations on two larrys with unaligned labels.

For the most part larry acts like a Numpy array. And, whenever you want, you have direct access to the Numpy array that holds your data.

Larry adds the convenience of labels, provides many built-in methods, and let’s you use your existing array functions.

Tabular

Tabular data can be easily represented in Python using the language’s native objects — e.g. by lists of tuples representing the records of the data set. Though easy to create, these kind of representations typically do not enable important tabular data manipulations, like efficient column selection, matrix mathematics, or spreadsheet-style operations.

Tabular is a package of Python modules for working with tabular data. Its main object is the tabarray class, a data structure for holding and manipulating tabular data. By putting data into a tabarray object, you’ll get a representation of the data that is more flexible and powerful than a native Python representation. More specifically, tabarray provides:

• ultra-fast filtering, selection, and numerical analysis methods, using convenient Matlab-style matrix operation syntax
• spreadsheet-style operations, including row & column operations, ‘sort’, ‘replace’, ‘aggregate’, ‘pivot’, and ‘join’
• flexible load and save methods for a variety of file formats, including delimited text (CSV), binary, and HTML
• sophisticated inference algorithms for determining formatting parameters and data types of input files
• support for hierarchical groupings of columns, both as data structures and file formats

Note to NumPy Users: The tabarray object is based on the ndarray object from the Numerical Python package (NumPy), and the Tabular package is built to interface well with NumPy in general. In particular, users of NumPy can get many of the benefits of Tabular, e.g. the spreadsheet-style operations, without having replace their usual NumPy objects with tabarrays, since most of the useful functional pieces of Tabular are written to work directly on NumPy ndarrays and record arrays.

Tabular requires Python 2.5 or higher but is not tested with Python 3k. Tabular requires NumPy v1.3 or higher. Any earlier version WILL NOT WORK.

Tabular is in beta!

See documentation at http://www.parsemydata.com/tabular.

Allegro (Allegrodev.com)

Credit Risk – Allegro’s credit risk solutions provide functionality to connect to major credit agencies to download the most current counterparty credit data available, thus reducing the time-intensive and error-prone task of getting the credit data into the system.  The solutions provides a wide variety of credit analysis tools to evaluate not only the current state of a counterparty’s credit status and liquidity, but to look forward in time as well to determine the potential future counterparty credit exposure risk. Additionally, the robust counterparty credit management tools interact seamlessly  with trade execution and contract management to provide a much more accurate and efficient credit management processe and much better credit risk data to integrate with the overall risk evaluation of the enterprise.

Hedging Risk – Allegro’s hedge accounting and reporting capabilities further enable compliance with support for FASB 133, FASB 157, IAS 39, FAS 159, FAS 161 and many other global regulatory standards.  Key areas include Financial Instrument Management, Currency Management, Interest Rate Management, Fair Value Management, Fair Value Option Management and Disclosure Management.  Providing robust and advanced analytical tools for analysis and reporting, it provides the control, accuracy, timeliness, and security that senior management requires to certify financial reports with confidence.
Financial Visibility- Allegro’s streamlined, highly automated processes for managing risk, price, and supply and demand curves allow managers to develop hedging strategies and estimate asset value. Drill-downs enable transparency of risk exposure at any level across physical and financial positions. Risk functionality is seamlessly integrated with physical logistics, enabling users to examine risk down to specific product deliveries. Position reports present managers with real-time reporting of exposure, limit management and profit and loss across the entire portfolio.

Real-Time Analysis – Allegro’s solutions allow risk managers to understand the impact of incremental positions on a portfolio, and by utilizing scenario analysis gain insight into potential risks and rewards. Allegro’s risk capabilities include real-time simulations, volatility analysis (all Greeks) and pre-trade credit verification. The risk simulation “what-if” tool allows companies to more accurately determine how potential market changes will affect various types of risk exposure. By employing simulation, users can identify potential risk behavior (under various market conditions), anticipate market moves, and act accordingly to achieve superior portfolio performance.

Advanced Trading Environment – The Allegro platform offers traders and managers the flexibility to customize dashboard views and reports (by trader, portfolio, etc.) with live information. Workflow and alert engines allow for dynamic reports and streamlined processes providing the right information at the right time.  The platform delivers out-of-the-box web services with connectivity to external data sources, including pricing sources (LIM and Platts) and commodity exchanges like ICE and NYMEX.

Imagine Software (www.derivatives.com)

Today the Imagine Trading System is available as a traditional in-house enterprise installation or as an ASP-based service, Derivatives.com. The former is the option for clients whose business requirements demand the use of proprietary valuation models.

The ASP option also eliminates the significant internal IT support requirements normally associated with robust, institutional-grade financial software.

Cross-Asset Support
The Imagine platform easily handles all major asset classes, spanning:

Equities
Equity derivatives
FX (Foreign Exchange)
CommoditiesFixed income
Credit
Swaps
Exotics

Multi-Strategy Support
Equally important, Imagine supports all of today’s major trading strategies, including but not limited to:

Equity long/short
Exotics
Volatility/dispersion
Convertible arbitrage
Statistical arbitrage
Distressed
Fixed-income arbitrage
Global macro

Sungard (Sungard.com) – APT Market Risk Models

APT delivers multi-factor models that use a pure statistical methodology based on market data, for factor definition.  More than 50 multi-factor risk models are available in single country, regional and global varieties, including dedicated emerging markets and Arabian markets models.  These models are used to estimate risk profiles for a multitude of asset classes including equities, bonds, currencies, indices, commodities, derivatives and funds.

Features:

Robust Modeling
Objectivity that helps ensure no poorly specified or arbitrary factors are included; while ensuring stability and responsiveness when you need it most. Regular updates make certain the models self-adjust to any market movements. This allows for a more rapid response when attributing the sources of risk to your portfolio. The use of sound statistical theory does not assume the normal distribution to ensure allowances for extreme events in “fat tails.”

Extensive Asset Class Coverage
Instrument coverage extends over a multitude of asset classes including equities, bonds, currencies, indices, commodities, REITs and funds.  In addition, Monte Carlo techniques can be implemented to the models to support derivatives.

Range of Forecast Horizons
Models are available in both short-term and medium-term risk forecasts.

Expertise
APT multi-factor models are regularly tested by a team of experienced professionals, improving on existing methods and innovating where necessary.

StatPro (www.statpro.com)

StatPro’s risk management module is a ‘firm-wide’ portfolio risk assessment tool for

integrated market and credit risk
counterparty risk
liquidity risk

More than 250 pricing functions

The multi-asset class portfolio risk management software covers the entire span of risk factors and currently delivers more than 250 pricing functions; a number that is growing continuously.

Bloomberg (http://www.bloomberg.com/professional/portfolio_risk_management/)

Bloomberg’s Portfolio & Risk Analytics solution streamlines the portfolio management process by integrating essential portfolio tools into one user friendly, customizable and intuitive interface. Precisely measure your portfolios against benchmarks, apply a range of attribution defaults, identify out- and under-performance, and gain unmatched transparency into portfolio risk.

Features include:

• Attribution
• Ex-Post Risk
• Ex-Ante Risk
• Real Time P&L
• Value-at-Risk
• Historical Performance
• Characteristic Reporting
• Scenario Analysis
• Available for BlackBerry® and iPhone® and iPad

Hyper Rig (www.hyperrig.net)

Risk Data Hyper Cube

In order to be able to perform consistent risk calculations and aggregation across multiple dimensions, Hyper Rig automatically generates a n-dimensional hypercube: users can specify and edit reporting hierarchy. The reporting hierarchies can be tree structures for books as well as flexible portfolios the system can accommodate a comprehensive Multi-Asset Portfolio

Indent  FX
Indent  Interest Rate Derivatives
Indent  Equities
Indent  Credit Derivatives
Indent  Energy

examples of dimensions include: asset type, hierarchy of book/portfolio, risk factor, counterparty for OTC products, etc.

Hyper Rig’s Risk Engine

Hyper Rig’s Risk Engine allows users to visualise aggregated risk positions across any dimension (whether market risk, counterparty risk, credit risk, liquidity risk) to reveal comprehensive risk exposures taken by individual portfolio managers or business units and understand how those risks are interlinked. Users can then slice and dice data from enterprise-wide aggregated risk exposures down to trade level for a timely monitoring of risk to:

Monitor Real-Time P&L.
Compute and maintain an accurate view of risk exposures and sensitivities.
Carry out what-ifs analysis in real-time at trade/book/portfolio/business line level according to user-defined scenarios in order to review the impact of a particular trade or group of trades on the key sensitivities of a book.
Perform for example a full Monte Carlo VaR in minutes instead of hours to assess how a trade will impact portfolio risk positions on a pre-trade basis.
Estimate current as well as potential exposures on a pre-trade basis.
Stress test any risk factor of combination of risk factors (any variable whose value will affect the valuation of the book such as interest rate, index, oil price etc.) and assess in real-time the impact of any market event.
Monitor trading limits in Real-Time.

Algorithmics (www.algorithmics.com)

Knowing your risk is a competitive advantage that companies can use to support informed decision making and business growth. Since 1989, Algorithmics’ services team has helped more than 300 of the world’s leading financial institutions realize their goals with a broad range of risk management services, backed by one of the largest research and development teams in the industry.

Portfolio Science (www.portfolioscience.com) – RiskAPI (Risk Application Programming Interface)

For Hedge Funds

Set up a risk management infrastructure for your fund in record time and minimum cost. Generate robust risk reports and perform dynamic exposure simulations. Meet both investor and internal requirements for risk control. Avoid costly back office integration.
For Portfolio Managers

Understand exposure. Benchmark risk to hundreds of indexes, commodities, or ETF’s, domestic or international, single, or multi-currency. Run simulations over flexible data sets to gauge the risk/reward of your holdings.
For Prime Brokers

Measure your client-wide exposure. Provide your clients with dynamic risk analytics and robust reporting. Users can instantly generate dynamic Value at Risk (VaR), beta, correlation, and more. Deploy RiskAPI to every fund client, under your own brand, at minimum cost.
For Fund Administrators

Provide your clients with a multi-asset, multi-currency risk reporting as a value-added service. Instantly add a whole spectrum of risk calculations to your existing account reporting framework. Paired with RiskAPI, the result represents unprecedented value to both current and future clients.
For Execution Providers

The real-time nature of RiskAPI allows you to provide your clients with position risk no matter how many trades they do every day. Users can instantly generate dynamic Value at Risk (VaR), beta, correlation, and more. Integrate sophisticated risk analytics into your pre-trade functionality, a winning combination.

Risk Data (www.riskdata.com)

Asset Manager: Risk Ticker Station

Riskdata’s Risk Ticker Station enables users to benefit from the power of a full reevaluation Monte Carlo Risk Engine INSIDE your favorite Front/Middle/Back Office application.

Hedge Funds Manager Risk Control: HEDGiX

HEDGiX was specifically developed for Hedge Funds, Riskdata’s HEDGiX is a complete risk management solution particularly adapted to dynamic investment strategies. The range of practical risk reports produced by HEDGiX matches both in-house active management needs and the growing demands of institutional investors for clear, relevant, and comparable risk analysis.

Fund of Funds Manager Risk Control: FOFiX

Developed exclusively for Funds of Hedge Funds, Riskdata’s FOFiX is a complete performance and risk profiling solution combined with a risk based portfolio construction tool. It enables FoHF to structure and apply a risk budgeting approach, consistently throughout the entire investment process. FOFiX uses methodologies that account for Long-term, short-term, non-linear and extreme behavior of Hedge Funds strategies.

HedgeFund.net (HFN) - delivers the database of active hedge funds, funds of funds and CTA/Managed Futures investment products. HFN’s database is easily exportable to Excel, Access, .CSV and SQL via HFN DataExport tool. HFN and PerTrac Financial Solutions offer a jointly developed product, PerTrac Online (PTO) to provide advanced analytics of the hedge fund related data. Please visit www.HedgeFund.net for more information.

BarclayHedge.com – contains comprehensive and up-to-date
hedge funds, funds of funds, single managers and much more. Similar to HFN mentioned above offers downloadable data integrated with online reports. Offers a free downloadable sample of its data. BarclayHedge also provides analytics tool called DataFinder Pro which comes free with the purchase. Please visit www.BarclayHedge.com for more information.

Hedgeco.net – delivers information to pension plans, family offices, consultants, funds of funds, banks, insurance companies, foundations, endowments and qualified private investors. Provides hedge fund data analytics. Please visit www.Hedgeco.net for more information.

HedgeFundResearch.com – provides a hedge funds database, fund of funds database, UCITS III compliant hedge fund database, dead hedge funds database. Offers database packages for investors and managers. Offers consulting and analysis services. Please visit www.HedgeFundResearch.com for more information.

EurekaHedge.com – Eurekahedge is the world’s largest alternative investment funds research house, specialising in hedge fund databases. An independent consultancy and advisory body in the Asian and European hedge fund markets, Eurekahedge maintains a hedge funds list of 24311* funds, across all strategies and asset classes.
As the only Asian located and focused service provider, Eurekahedge has a database covering over 400 Asian portfolios, improving access to regional hedge funds. With the combination of a strong database and independent research team, Eurekahedge is able offer unrivalled ongoing and objective investment advice to potential investors in Asian hedge funds. Please visit www.EurekaHedge.com for more information.

Morningstar Hedge Fund Database - Morningstar’s hedge fund database contains data on 8,000 funds available to both U.S. and international investors. Morningstar brings the same transparency to the hedge fund industry that it brings to other investments, and is committed to helping investors and financial professionals properly evaluate hedge funds. And, this data is available in Morningstar products for quick, easy analysis and reporting. Please visit http://corporate.morningstar.com/us/asp/subject.aspx?xmlfile=545.xml for more information.

Lipper TASS Hedge Fund Database – The Lipper TASS database is an indispensable resource for institutional asset managers, high net worth investors and consultants who monitor the global hedge fund industry. Subscribers of Lipper TASS now benefit from Lipper’s data-collection and analysis expertise. Lipper TASS Hedge Fund Database global database has been a reliable source of timely, high-quality hedge fund data for over 20 years. Lipper TASS offers quantitative performance data on over 6,300 actively reporting Hedge funds and Funds of Hedge Funds, plus over 7,000 graveyard funds that have closed, liquidated or stopped reporting for any reason. Please visit http://www.lipperweb.com/products/LipperTASS.aspx for more information.

EuroHedge Database – offers an extensive database of European single-manager funds. Contact details of funds are listed, many of which include full performance data. You can create your own ‘watch list’ of funds in a bespoke portfolio, by adding individual funds or entire lists.  Data is download-able. Please visit http://www.euromoneyplc.com/product.asp?PositionID=& for more information.ProductID=7356 for more information.

Preqin – In 2007 Preqin launched its hedge fund services to meet the increase in institutional investment in the asset class over recent years. Preqin research indicates that institutional investors are making up 70% of an average hedge fund client base, and are becoming a more sophisticated, important source of capital for the industry. Hedge funds therefore identify institutional investors as major targets for obtaining capital. Please visit http://www.preqin.com/section/hedge-funds/2?rid=24 for more information.

HedgeFundContacts.com – with access to 3,500+ funds and more than 8,000 individuals, HedgeFundContacts.com’s Firm Search and People Search databases supply alternative investment contact information that is unmatched in terms of ease of use and industry coverage. Please visit www.hedgefundcontacts.com for more information.

HedgeFundLaunch.com – Hedge Fund Launch is happy to announce that we are now providing start up and emerging hedge fund managers a chance to add their fund information for inclusion in our exclusive database.   Our database comprises data from start up and emerging hedge fund managers only – and will be available for purchase in the near future. Please visit www.HedgeFundLaunch.com for more information.

NewHedge.net Hedge Fund Investors Database – provides specific details for over 7,000 hedge fund investors, covering 22,000+ contacts worldwide who currently invest in hedge funds, and fund of funds. Each contact includes: listing name, investor type, geographic location and main contact person within firm. Please visit www.newhedge.net for more information.

In a nutshell, it is an open source library for quantitative finance, written in 100% Java, which aims to ‘port’ QuantLib from C++ to Java,
keeping exaclty the same API (as much as possible) and performing very closely to the C++ implementation (QuantLib).

A recommended approach for interviewing candidates is to draft a specific list of questions before the interview begins that are pertinent to the position that needs to be filled. These questions should be based on the candidate’s QA accomplishments and skill sets as documented in the resume. Below are some sample questions that can help demonstrate a candidate’s experience in the areas of quality assurance, and also their creativity and ability to comprehend basic testing concepts. The questions below can serve as criteria to screen out inexperienced candidates during the interviewing process.

Describing Testing/QA Terms

An experienced testing candidate with several years of experience will exhibit knowledge and understanding of well-established testing principles and testing terminologies. It behooves the test manager to ask the testing candidate to describe these concepts. Some suitable examples would be:

What is the objective of a peer review?
What is the Unified Modeling Language?
What are the components of a test plan?
What are the benefits of automated testing?
What are the benefits of Configuration Management?
What are the characteristics of a good test requirement?
What is a requirements traceability matrix is and why it is necessary?
What is the criterion for composing a test readiness review list (TRR)?
Provide descriptions for testing approaches (i.e. white box versus black box, etc).

The test manager needs to ascertain if the testing candidate is familiar with industry accepted terminologies that are commonly used within the project where the tester is being considered.

Thinking On Their Feet

In addition to understanding the testing requirements, a good testing candidate should have creativity and ingenuity when testing a software application. A tester should always be alert to potential scenarios that could cause an application to fail or yield defects and/or errors; even if such scenarios are not documented or presented in requirements. A thorough tester executes a particular test scenario with different sets of data, and conducts boundary testing to ensure that an application would not be deployed into production with overlooked problems.
A suggested question to discern a candidate’s testing meticulousness is to have a candidate provide use cases and test cases for a commonly used machine, or a self service application. An example would be: what test cases and use cases can the candidate think of for operating a beverage dispensing machine, or for purchasing books via a website. The candidate should generate an extensive list of test cases and use cases for the two aforementioned examples.

Development of Test Scenarios and Test Scripts A well written test scenario has information about: pre-conditions, post-conditions, traceability to a requirement, description, identification of authorship, a peer review and sign off section, roles to be tested, etc. A test script or test procedure on the other hand has detailed test steps with valid data values and expected results for each test step. In addition, a test script provides information about the test execution results for “passes/failures” and mapped requirements for the test step.
The test manager can have the tester provide information as to how test cases and test scripts were documented at the previous project. What level of detail was presented for the test cases and test scripts? What exactly was documented for the test cases and test scripts? Another suggestion is to have the tester send a sample test case and test script that he/she documented at a previous project for review. A well-documented test case and test script will demonstrate the tester’s attention to detail.

Life Cycle Methodologies

An experienced tester should have experience working with one or more IT methodologies such as waterfall spiral, evolutionary, incremental, rapid prototyping, etc. Some methodologies are more appropriate when requirements are well known/defined, or when requirements are not well known, or when the project has high risks, etc. The tester vying for the position should understand what the differences are between the main software development lifecycle methodologies. The test manager can present the tester with different hypothetical scenarios of IT projects and ask the tester what methodology would be most fitting based on the presented scenario.

Test Procedures and Test Standards

What testing standards and procedures the candidate has been exposed to is of paramount importance in determining whether the candidate would either be a good fit or adapt to a new testing environment. Did the candidate come from a regimented and disciplined type work environment like a CMM work environment with repeatable and defined processes? Or, did the candidate work in a chaotic test team that did not have any standards, procedures or defined processes?

Testers that come from a loose testing environment sometimes have difficulty adjusting to regimented testing environments that have defined processes and standards for things such as: lessons learned, test plans, naming standards, version control, test case templates, test execution matrix, test logs, test folders, reporting of test results, etc. Conversely, a tester that follows strict testing standards and procedures may struggle in a work environment that does not have defined processes, well documented scenarios, no structure for test cases, or templates for test scripts, etc. Questions that identify what the candidate’s experience is with testing standards and procedures are critical for assessing how well the candidate would adjust to the current QA environment.

Defects

It’s advisable to learn what sorts of defects a tester has identified and reported on in previous projects. Based on the candidate’s answers, the test manager can learn whether the tester focuses on cosmetic/minor defects or defects that are show stoppers and would have caused havoc for a released or production deployed application.

The test manager can have the candidate expound on a significant defect that was discovered and, what the impact of the defect would have been had the candidate not caught it. In particular, the test manager can focus on the tester’s approach for identifying the defect and how the tester re-tested the defect subsequently leading to the closing of the defect.

Test Script Automation Many candidates list in their resumes that they have experience with various automated testing tools. However, when confronted with technical questions about their experience with automated testing tools many candidates respond that their experience with automated test tools is limited to capturing and playback. Candidates that manifest to have only recorded and played back test scripts are in fact devoid of significant experience with automation test tools. Even candidates who claim to hold certifications with test tools should have their background probed and examined with technical questions.

The test manager should find out whether the candidates understand what a data driven test script is, how to create a parameterized test script, and why it would be necessary to construct a data driven script. The test manager should also ask the candidates to explain what data correlation is, how to create a driver script, how to synchronize scripts, why it is necessary to synchronize scripts, and how to report the execution results from a test script. Other key questions would be techniques for identifying and verifying recorded objects including providing examples of object’s attributes that can be verified, how to create checkpoints, how to debug an automated test script, and in what modes a recorded script can be re-played. The candidate should also know how to map and learn custom objects that the recording test tool does not identify or recognize.

The test manager can present more technical questions based on the nuances of the recording test tool that is present at the project site. The main objective of technical questions is to ensure that a candidate has experience that goes beyond merely recording and playing back scripts.

Rather than limiting this to a formally technical process used to investigate and measure quality of the developed computer software in terms of correctness, completeness and security, it is imperative to test the software components and the software as a whole for integrity, capability, reliability, efficiency, portability, maintainability, compatibility and usability of the software. This is particularly important in an operating environment where it is planned to be implemented, and testing will need to adhere to the technical requirements as described under ISO standard ISO 9126.

Test Management

Planning a Test Project

A test project includes the creation of a test plan, collecting test scenarios, writing test cases, executing test cases, evaluating and reporting the test results and managing the software testers.

It is also imperative to ensure which items are in the scope of testing and which are out of scope, maintaining scheduling and training. We need to establish a strategy to be followed in each of the testing phases. All of these phases come under test management.

Managing Test Data

Effective testing requires the presence of adequate and suitable test data. Where test results are specified, it is presumed that the inputs will be consistent from text execution to execution. In situations where automated testing methodologies are employed, a standardized test bed of data may be required to facilitate the automated assessment of “passed”.

Depending on the test being executed, test data may be copied from a production system, or may be developer generated, or a combination of both. In situations where extreme testing precision is required, the test data may be specifically created, saved and copied into the environment at the outset of every testing cycle to ensure a consistent test execution.

Test data needs to be considered when developing some of the larger tests (User Acceptance Testing, Regression Testing and System Testing). As plans are developed for test execution, the creation, modification or even restoration of test data are critical concerns that must be effectively managed.

Testing Methodology

The best approach is to conduct end to end testing. Testing methodology is applied in two major phases: Functional Testing and Non-Functional Testing.

But also the context where a system is deployed will influence the Testing Methodology. There will be a difference between tests for a commercial software package to be sold to consumers and embedded software or a server solution for a website of a one-off logistics solution.

Functional Testing

1. Unit Testing : Testing each component of the application separately.
2. System Testing : Testing the system as a whole.
3. Integration Testing : Testing the system with other interfaces.
4. Regression Testing : Testing changes made to the system and ensuring that new problems are not introduced as a result of the changes.
5. User Acceptance Testing : Testing the system to ensure it meets the user requirements.
6. Sociability Testing : Testing the system with other applications in the platform.
7. Security testing : Test the software from external damages it meets the user requirements.
8. Performance testing : Testing that is performed to determine how fast some aspect of a system performs under a particular workload.

Non-Functional Testing

1. Usability Testing : Usability testing focuses on determining if the product is easy to learn, satisfying to use and contains the functionality that the users desire.
2. Stress and Performance Testing : The purpose of this testing is to predict the system behavior and performance.
3. Accessibility Testing : to determine the extent to which the end user interacts with the application.

Automated Testing

One of the purposes of automated testing is to ease regression testing. Once the system is stable, then automated testing can be started.

Once automated tests exist, they can and should be used to help build better and more stable software ready for any manual testing and user acceptance testing.

Preferrably, automated tests can be run overnight on software builds, rather than the live version. This will help give a clearer indication to the stability of the build the following morning. Hopefully, leading to more time being available for the rest of the testing and development.

Test Framework

A set of ideas and tools that revolve around optimizing the test-effort. When implemented using the supporting tools, these ideas render the maintenance of the automated testing scripts or other entities easier.

JUnit Test Tool

The JUnit test tool is designed to assist in unit testing of java objects. It can be useful in testing a non-java environment (eg. by wrapping the test target inside a java object).

From Main Frame testing to ebusiness testing

* Testing the MainFrame System
* Testing the Client Server Application
* Testing the Internet Application
* Testing Web Services

Component Performance Criteria

Testing is a major part of agile development processes, but until recently, the vast majority of effort has been focused on functional testing. In the world of integrated systems, many projects are undertaken to speed up processes. Traditional development processes teach that optimisation should not be carried out until a performance problem is identified, thus aiding maintainability. Unfortunately, this has led to the postponement of performance testing until after integration testing. This can often make it too late to address fundamental problems as time is short or key resources have moved on to other projects.

Component Performance Criteria is about breaking down the performance requirement for the whole system and attributing it to individual components. Admittedly, it does not guarantee the absence of performance problems later in the project, but it will highlight if they are already there.

Example: if a system must process five transactions per second, and consists of 10 major components, and an individual component has been measured to take 0.5 seconds to process a single transaction and can only process one transaction simultaneously, a bottleneck is already present and resolution of the problem can begin immediately.

Optimally, performance testing should be carried out regularlly with the fewest potential issues. This is effectively done with automation.

Know The Client
The holy grail of any project, satisfy your client. For that you need to know, what is in his mind regarding his requirements. He is the best person to tell you, what they want. While skilled and experienced software engineers can fill in any incomplete, ambiguous or contradictory information.

Specification
Specification means precisely describing the software to be written in a rigorous way.

Software architecture
The architecture of a software system refers to an abstract representation of that system. Architecture is concerned with making sure the software system will meet the requirements of the product, as well as ensuring that future requirements can be addressed.

Implementation (or coding)
Reducing a design to code may be the most obvious part of the software engineering job, but it is not necessarily the largest portion.

Testing
Testing of parts of software is done by a software engineer to check out any bug or error.

Documentation
An important (and often overlooked) task is documenting the internal design of software for the purpose of future maintenance and enhancement.

Black-Box testing

* Examines the fundamental aspect of the system with little or no regard to the internal logical structure of the system.
* Demonstrates that software functions are operational and the input is properly accepted and output is correct produced.
* Integrity of the external information (eg. Database) is maintained.

BLACK BOX TESTING STRATEGY – would involve different types of Testing. When the Build is for the first time to test – we start with Validations and Verifications wrt the Scope/SRS document or Mapped Test Cases. We even start tracing the Traceability Matrix. We mark the observations as in Severity and Priority and publish a Testing Report. Next Build onwards we start with Smoke Testing Phase – if test cases report FAIL STATUS then we reject the build saying it Non Testable. If the cases are of PASS STATUS we start with System Testing, Include Regression Testing – verifying the old bugs and though Impact Analysis of each Bug. If new functionalities are there then we do Integration Testing. IN ALL the Builds we incorporate Usability Testing – checking UI, Messages, and other User perspective points. We incorporate Database Testing – through SQL Commands to check the Data Driven Scenarios.Last but not the least cover the Security Testing phase – especially in cases of web applications.

White-Box testing / Glass-Box testing

* Examines the fundamental aspect of the system with complete information and access to the internal logical structure, code and algorithms.
* Logical paths are tested. Test cases which tests the loops (DO LOOP) and conditions (IF STATEMENTS) are explicitly designed and used to test the logical paths.