STATISTICA Blogs

Recently, I sat down with Gary Miner to understand why and how he uses STATISTICA. 

• Tell us about your background (career, education, etc.). I was always interested in math and science. I became very interested in genetics during college at Hamline University and went on to complete a Master’s in Physiological Genetics at the University of Wyoming, and then continued in Biochemical Genetics at the University of Kansas and the Jackson Laboratory in Bar Harbor, Maine. This was followed by Post-Doctorals in Behavioral Genetics at the University of Minnesota, which shifted into a research professorship studying the biochemistry of schizophrenia and Alzheimer’s disease. Later, I completed another Post-Doctoral at the University of Iowa in Psychiatric Epidemiology. Alzheimer’s disease became my prime research area, and about 30 years of my life has been devoted to this disease. During this research career, I used SAS, SPSS, and BMDP software. At the time, I liked BMDP the best, as it was “modular” for each statistical analysis. Then in the late ‘90s, I discovered STATISTICA, which was developed “in my back yard” here in Tulsa, so when I was looking for a “business career” after the years of medical research, I decided that working at StatSoft would be a perfect “semi-retirement career” after having used statistics to analyze data during my 30-year research career. I soon discovered that STATISTICA was light years ahead of the competition in its user friendliness. I also discovered that STATISTICA’s graphing was absolutely “first class.” Of course, I’ve had great fun writing two books on Predictive Analytics as adjunct projects during my time at StatSoft.
   
• How do you use STATISTICA? Which modules? I primarily use STATISTICA to analyze medical research data, keeping my fingers in “academia” by serving as “community faculty” for the IHI Medical Residency program in Tulsa. The IHI is a family practice residency. The second-year residents are required to design and complete a research project using statistical analysis. Since these are family practice residents, their interests are broad and varied, so we work on all kinds of research problems from very basic science to sociological instruments, medical care delivery business research areas, and even genetic-psychological-psychiatric research. I primarily use Basic Stats, the graphing capabilities, and GLM and Discriminant Analysis with the medical residents; however, about 70% of our medical resident studies also use Predictive Analytics, e.g., the data mining algorithms, as these are what are really needed to generate more accurate predictive models required to make medical knowledge and medical care delivery more precise and cost-effective.
   
• Why did you choose to use STATISTICA over another product? Any statistician/data analysis expert needs to use more than one software package, as each will have some strengths. However, most statisticians will gravitate to one as their “main foundation.” STATISTICA is about all I use now, as it has open architecture, a simple customization language, and greater flexibility than I can find in any other package on the market.
   
• Did STATISTICA give you the solution you were seeking? Yes. It gave me the best solution and the most algorithms, with three different approaches for conducting Predictive Analytics studies. I can choose to use the Data Miner Recipes UI when I want to be guided through the steps, the more comprehensive  Data Miner Workspace, or when I need to make more tweaks to the parameters I can use the Interactive Data Miner module format
   
• What are STATISTICA’s strengths? STATISTICA has flexibility, open architecture, and simplicity of use, yet also has the ability to fully serve the needs of complex business organizations via the Enterprise format of the software.
   
• Would you recommend STATISTICA to a colleague or friend? Why? I always recommend it to my colleagues because of the simplicity yet great power and accuracy as validated by its benchmarks.