Getting Started with Statistics – a key part of Business Intelligence

A key part of our work in Business Intelligence is the analysis of data and the branch of mathematics that we employ for this is statistics.

We analyse data to understand our business better and when analysing the data our main objective is to pick out the reality from the background noise. For example when looking at weekly sales figures we see that the number of items sold varies from week to week, some weeks more, some weeks less. The challenge is to understand whether this change is just chance or there is something driving the change – a marketing campaign, or favourable comments on social networks.

By understanding our business better we can make decisions to improve our business, for example, if we can prove that a marketing campaign is not making any difference then we can save some money but cancelling the campaign. But being able to “prove” that a campaign is, or is not, making a difference we need to use statistics.

Recently Nathan Green started a new series of articles on statistics in the UK’s Guardian newspaper. So far I have found these articles to be very well written and provide a good introduction to statistical analysis, providing clear explanations for tricky concepts such as the difference between mean, median and mode and what is a p-value by using every day examples.

These articles are looking like a great way to get started with understanding statistics and so I would recommend you check them out. And if you already are adept with statistics then these still provide a good source of interesting material!

Enjoy,

Al

SAP Data Services: 10 Reasons for never using Global Variables (and one where you can)

In software programming it is widely held that using global variables or other constructs to maintain global state is a poor programming practice[1] and many of those reasons can also be applied to development of jobs in SAP Data Services.
The following is my list of 10 reasons for why you should never use global variables,

  1. Tight coupling between separate parts of a Data Services job
  2. Hidden dependencies between separate parts of a job
  3. Lack of isolation for unit testing components of a job
  4. No access control – everything in a job has full access to a global variable
  5. Global variable values can be set at runtime
  6. Lazy reuse of global variables
  7. Global variables depend on the context in which they are being used
  8. Local variables can have same name as global variables
  9. Global variables discourage code reuse
  10. Global variables restrict configuration management

These are further explained below and since I’m a never-say-never type I’ve also included one situation where you can use a global variable.

Read the rest of this entry »

Overview of Moving Averages

Introduction

A moving average is a simple technique for smoothing random data. Most often we find moving averages to analyse movement of stock prices but we also see them in other areas of business and data analysis.

This is the first part of a series of two articles. This article discusses what are moving averages and how they are calculated. The second part then looks at how to implement moving average calculations in SAP BusinessObjects Web Intelligence.

If you already understand moving averages you can skip to the second article on how to implement in Web Intelligence.

Read the rest of this entry »

Calculating Moving Averages in Web Intelligence

Introduction

The previous article looked at what moving averages are and how to calculate them. This article now looks at how to implement these in Web Intelligence.

The formula used here are compatible with the XIr3 version of SAP BOE however some formula may work in previous versions if available. We’ll begin by looking at how to calculate a simple moving average before looking at weighted and exponential forms.

Read the rest of this entry »

Creating Histograms in Web Intelligence with Adjustable Bucket Sizes

Introduction

This article looks at how to create a histogram in Web Intelligence and makes use of input controls to allow the user to automatically adjust the histogram’s bucket size.

Histograms is a charting technique used to analyse the distribution of a set of data for example a business analyst may use a histogram to analyse the range of ages of the employees of a business. Or in manufacturing you can use a histogram to analyse how stable a process is.

A histogram is one of the seven basic tools of quality and is also used in Six Sigma and other business improvement methodologies.

Read the rest of this entry »