### Individual Blogs

 View All Recent Entries Analytically Speaking (6) Data Visualization (31) Enterprise Analytics (6) Numbers in the News (13) Practical Analytics (1) Predictive Analytics (1) Project Management (10) Random Data (2) Round Hole, Square Peg (23) STATISTICA (110) STATISTICA How-To (95) STATISTICA Press Releases (75) STATISTICA-ly Speaking (6) Statistimix: The Urbane, Arcane, and Inane (10) Talking Code (4) Thursdays with Therman (4)

# STATISTICA News and Blogs

## Statistica thought leaders in the mix

by STATISTICA on Tuesday, November 11, 2014 10:55 AM

Statistica subject matter experts keep busy, staying abreast of current trends with big and small data, predictive software and real-world analytics solutions. And they frequently comment in industry publications, professional forums and blogs. Here are a few of their recent articles for your reading pleasure...

## Line Plot or Scatterplot?

by STATISTICA How-To on Wednesday, October 29, 2014 12:31 PM
A line plot is usually used for visualizing a sequence of many numeric values (e.g., stocks over days, temperature across regions, etc.) In contrast, a simple scatterplot is widely used to explore the relationship between two numeric values.

You can use Statistica to create both types of graphs, which sometimes look very similar. However, there is a big difference in how Statistica treats the x-axis variable (i.e., the variable used to produce the horizontal axis) when producing each of these two types of graphs. Knowing this difference can help you choose the right tool for your needs.

Statistica senior analyst, Wendy Xie, demonstrates the difference in this month's article...

## Controlling weather with a hammer? Trends in predictive and advanced analytics

by STATISTICA on Thursday, September 11, 2014 11:02 AM

StatSoft co-founders Thomas Hill and Paul Lewicki were invited to contribute a forward-looking article to the 20th anniversary issue of Scientific Computing World, a publication that has reviewed the Statistica product several times over the years.

With 30 years of the Statistica legacy behind them, and now as executive directors with Dell Software Group, Hill and Lewicki took this opportunity to consider what they believe are the top changes and trends driving the current and next waves of analytics technologies...

## How to Unstack/Stack Data in Statistica

by STATISTICA How-To on Wednesday, September 10, 2014 2:15 AM

Statistica offers options to perform “unstacking” or “stacking” of data files to make them more suitable for statistical analysis. The general operation or restructuring of data is similar to a cross-tabulation of values (given certain categorical variables).

These operations are especially useful when accessing data from databases where individual data values are identified by case and variable IDs, and where each unique combination of these IDs is contained in a new row of the data file.

In this how-to article, Krista Beasley, Dell Statistica's Technical Support Manager, walks you through some simple scenarios...

## Dell Software Execs on Analytics, Data Mining, and Statistica future

by STATISTICA on Wednesday, August 06, 2014 1:40 PM
In June, KDnuggets.com's Dr. Gregory Piatetsky published a summary of 2014 KDnuggets Annual Analytics/Data Mining Software Poll results. In his accompanying commentary, he observed, "Statistica's share has now declined for 2 years in a row (from 14% in 2012 to 9% in 2013 and 1.7% in 2014). With the recent acquisition by Dell, it seems likely that Statistica will continue to lose market and mind share."

This comment, of course, got our attention...

## Why the best experts can’t always explain what they’re doing

by STATISTICA on Tuesday, August 05, 2014 1:25 PM

An engineer once told me this joke: A spider asked the centipede, “You have so many legs; how do you coordinate them all to walk?” The centipede pondered this question for a while, then a while longer, and longer; and the centipede never walked again.

Many industries are buried in data to the point that it overwhelms the traditional approaches to process monitoring and root cause analysis performed by process experts. Yet, human process experts can often develop a “sixth sense” about even the most complex processes...but, like the centipede, they may not be able to provide any insights on how they accomplish this feat...

## Rescuing Businesses from the DIY Big Data Vendor Trap

by philler on Tuesday, July 29, 2014 2:47 PM

Recently, I received a product positioning template to be completed for our advanced analytics platform, STATISTICA. When I came to the section where I was to describe the "industry problem," I was stuck, because every industry has its own kinds of problems that can readily be solved with analytics.

And then it finally hit me: there is at least one common problem that every business must face in every industry...

## Predicting Best Treatment for an Individual Using Weight of Evidence (WoE) and Automatic Binning of Continuous Variables, Leading to Decisioning

by STATISTICA on Wednesday, July 09, 2014 7:36 AM
Guest-authored by Linda A. Winters-Miner, Ph.D., of Southern Nazarene University in Tulsa, this mini-tutorial comprises excerpts of two tutorials from a book to be released this summer titled Practical Predictive Analytics & Decisioning Systems for Medicine by Winters-Miner, Bolding, Goldstein, Hilbe, Hill, Nisbet, Walton, and Miner (2014).

The first part is the introductory text from a tutorial (F-2) that concerns weight loss, which was authored by Winters-Miner (2014). The second part is a tutorial (J-2) that concerns the medical condition Disseminated Intravascular Coagulation (DIC), which was authored by Khichi, Parsons, and Winters-Miner (2014).

It is hoped that, by combining some aspects of both tutorials here, this mini-tutorial will...

## How to Group Data into Specified Percentile Categories using STATISTICA Visual Basic

by STATISTICA How-To on Tuesday, July 08, 2014 10:51 AM

Sometimes, STATISTICA users want to group the data in a spreadsheet variable according to percentile categories. For example, I might want to know which cases fall into which quartiles. It would make sense to sort the data, as well. It is fairly easy to do this using an SVB macro, as shown in this example...

## Why predictions go wrong – and how to prevent it

by STATISTICA on Tuesday, June 10, 2014 5:31 AM
A few years ago, I received a monthly mail-order catalog featuring various pieces of self-defense weaponry, including things like brass knuckles, nunchucks, assault knives, throwing stars, and other combat fighting “systems.”

As anyone who knows me can attest, this was clearly an example of market segmentation and propensity modeling gone amok—and a waste of money on the part of the company that sent me the catalog...
Content