« Home | Welcome BIWA » | Merry Christmas, Happy New Year, and a Poll » | Announcement: Oracle Data Mining Consultants Partn... » | Free Webinar: Competing on Analytics » | Oracle Data Mining in Argentina » | Time Series Revisited » | Oracle Data Mining 10gR2 Code Generation Release N... » | KDD 2006 - Day 2 » | KDD 2006 - Day One » | New Code Generator for Oracle Data Miner » 

Monday, February 12, 2007 

New Oracle Statistical Functions Page

OTN has a new page (link) describing the statistical functions in the Oracle 10g Database. These functions are available in all versions of the database at no extra cost. Features include:

  • Descriptive statistics
  • Hypothesis testing
  • Correlations analysis (parametric and nonparametric)
  • Ranking functions
  • Cross Tabulations with Chi-square statistics
  • Linear regression
  • ANOVA
  • Test Distribution fit
  • Window Aggregate functions
  • Statistical Aggregates
  • Pareto analysis
  • LAG/LEAD functions
  • Reporting aggregate functions
These statistical functions are also very scalable. In fact, I have heard that some companies are using these functions to compute statistics on massive amounts of data that cannot be handled by traditional statistical tools. Check it out!

Labels:

This is interesting development. I would love to hear Oracle's or Database Industry's general philosophy about Predictive Analytics/Data Mining/Whatever you may call it.
Curiousity arises from the fact that I worked for a few years in the database world (mainly oracle)as a programmer & then managing Datawarehouses, ETL, Reports, etc. Then recently started working in a 3rd party Analytics Services Provider outfit.. mainly a SAS shop.. we hardly talk about tables or databases here. Its all about getting the data in SAS files on our servers & then doing the requisite analysis. I dont need to reiterate that whenever you are looking for enterprise level analytics, robust models.. there isnt much to look beyond SAS.
Whats the database companies' take on this? They control the source. So, why let another player control the downstream piece of the pie? It might be a small pie as compared to the DB or DW market right now but with adoption of analytics gaining momentum, this is one area where companies cant afford to lag.
your thoughts?

- Amit

Thanks Amit for the comments. I agree with the points you made. The analytics market is small compared with DB and DW but there is great potential for growth. I believe, from the database vendor point of view, it is a strategic necessity to have analytics in the database. From the customer point of view it provides major benefits in speed, security, and resource savings (no data duplication).

I personally believe that analytics will transform the way we interact with databases and the types of applications db developers can write. An important requirement on this road is to make analytics easier to use and robust under different data quality conditions. This is a step that Oracle has been pursuing consistently over the past releases. Oracle Data Mining in 11g provides major improvements in these areas.

Post a Comment

About me

  • Marcos M. Campos: Development Manager for Oracle Data Mining Technologies. Previously Senior Scientist with Thinking Machines. Over the years I have been working on transforming databases into easy to use analytical servers.
  • My profile

Disclaimer

  • Opinions expressed are entirely my own and do not reflect the position of Oracle or any other corporation. The views and opinions expressed by visitors to this blog are theirs and do not necessarily reflect mine.
  • This work is licensed under a Creative Commons license.
  • Creative Commons License

Email-Digest



Feeds

Search


Posts

All Posts

Category Cloud

Links

Locations of visitors to this page
Powered by Blogger
Get Firefox