Social Network For Analytic Professionals
Daniel Kocis's Groups
-
LinkedIn
36 members
-
SAS Network
294 members
-
Platinum Members
96 members
-
Analytic Jobs
435 members
-
Web Analytics
298 members
Daniel Kocis's Discussions
two-stage models to improve response prediction
Replied Jul. 23, 2009
Daniel Kocis's Page
Profile Information
- Short Bio:
- But how is Multivariate different..REALLY !
Responsible for performance based optimizations that leads to the development of process, media, web, and reporting strategies of non traditional business metrics and key insights back to senior management. Defined program strategy, vision, and infrastructure requirements that internally integrates these business objectives into data driven decision support. Multiple assignments within the Financial Services / Banking Credit Risk / Consumer Retail / TV Advertising / DTC Pharmaceutical Industries that used my experience synthesizing disparate data and multivariate techniques to uncover leveragable opportunities while focused on profitable marketing summarized with analytical dashboards and visualizations.
Previous Clients: ditech.com (multi-channel media and call center optimization, risk based vintage reporting that highted waterfall effects of new acquisition by product line), TransUnion-Peoples Bank-Fleet-Mellon-Chevy Chase (risk based market size and new customer acquisition models), Citigroup (transactional information products), Bear Stearns (RMBS pricing models), MRM (Sprint TV ad optimization, Neupogen promotional/educational marketing), TribalMetrix (Humira.com branding and promotion, ING website analytics), mktg (event marketing optimization and new site rationalization), Cushman & Wakefield (Accounting Dashboards).
Online Toolkit – Google Analytics – GATC (ga.js) - Adwords - Keyword Tool – New Tracking API’s
Statistical Toolkit - SAS 9.2, EMiner 5.3, ODS, WEB Report Studio, Proc SQL with ODBC, COGNOS 7
Personal Toolkit - Disruptive Thought leadership about leveraging business analytics, predictive statistical models, customer segmentation CRM, marketing break-even analysis , PRIZM and Nielsen television media buys using transactional behavior with waterfall analysis and vintage reporting focused on maximizing ROI on new sales.
In this highly competitive environment I need to point out some differences that are really critical. I am focused on timestamped transactions from both WEBLOGS and OFFLINE Master files. For a Mortgage Originators' inbound call center where TV ADS and WEB paid and natural keywords drive all activity while under dramatic media budget cuts. For a Pharmaceutical companys' unbranded website gathering pre-qualified emails from patients, separating caregivers and professionals. For an entertainment star online retailing of songs, ring tones, perfume, and accessories. For a Liquor distributor holding 1,000 tasting events a week, where site profiles impact ROI. For RMBS Bond dealer that is quantitatively warehousing and pricing new deals using a bottoms up loan level methodology. For a 24 hour store selling shelf space based upon POS sales. For a website developer that confuses visitors with a bad landing page. For a CPC site that is losing money with huge Bounce Back Rates.
Common among these efforts is dynamic streaming data that is mapped with other disparate informations sources, then transformed, modeled, and dashboarded into ACTIONABLE Business Intelligence.
Yes, I use a BIG toolkit of everything expected but I am not selling SAS, SQL, COGNOS, ORACLE expertise per diem, but providing multivariate solutions based upon your data which are generally NOT apparent to you.
Waterfall analysis identifies the fall off turning initial transactions into sales. Who cares how many impressions lead to how many visitors, when its the email address or paid download that counts. Vintage analysis takes snap shot online reports and JOINS them into your companies master files which expects future sales and ties ALL activity of that customer together for a TRUE 360 View.
- My Website:
- http://www.multivariate.com
- Field of Expertise:
- Data Mining, Biostatistics, Finance, Marketing Databases, Operations Research, SAS, Web Analytics, Statistical Consulting, Medical Statistics, Artificial Intelligence
- Years of Experience in Analytical Role:
- More than most.. less then others !!
- Professional Status:
- Consultant
- Interests:
- Networking, New Venture
The art and science of mathematical investigation begins with asking, “What kind of data do I have? “, then “How do I re-represent these raw numbers into something more meaningful?” Datum refers to a single metric, usually a number while data are a collection of numbers and alpha-numerics. Before selecting a tool and building models, the investigator needs to create an extended database from which to calculate their models. Preliminary questions include:
·Is it static (such as a one-time survey) or dynamic (periodic refreshes) data
·Do they exist as a single table or spread across several tables
·If they do exist across several tables, what is the hierarchy of each table?
This chapter addresses several issues and provides many different examples of creating-transforming raw variables into potential quantitative indicators and metric’s.
This chapter will show how to:
·Use streaming data for nuclear and physiological investigations,
·Create unique time-dependent hierarchies from flat data,
·Transform simple counting statistics to increase the number of new variables,
·Constrict confidence intervals with additional transformed variables,
·Create complex physiological ratios to predict heart failure in children,
·Use demo/geographic/transaction data that define high valued clients,
·Calculate financial filters to identify n-period ahead closing stock prices,
·Create “Discovery Paths” within a website that reconfigures with SEO,
·Capture meaningful metric’s from NCSA raw web logs,
·Use “relativistic coding” in time series analysis,
·Aggregate the dynamic quality of streaming data to create new variables
·Quantify text-mining from organic keyword searches on the web,
·Define “high-valued” or “target” segments
·Re-aggregate the hierarchy of data into meaningful segments
·Create dimensions for CRM based upon simple sign-in data
·Look at data and their transformations in new and different way
·Is it static (such as a one-time survey) or dynamic (periodic refreshes) data
·Do they exist as a single table or spread across several tables
·If they do exist across several tables, what is the hierarchy of each table?
This chapter addresses several issues and provides many different examples of creating-transforming raw variables into potential quantitative indicators and metric’s.
This chapter will show how to:
·Use streaming data for nuclear and physiological investigations,
·Create unique time-dependent hierarchies from flat data,
·Transform simple counting statistics to increase the number of new variables,
·Constrict confidence intervals with additional transformed variables,
·Create complex physiological ratios to predict heart failure in children,
·Use demo/geographic/transaction data that define high valued clients,
·Calculate financial filters to identify n-period ahead closing stock prices,
·Create “Discovery Paths” within a website that reconfigures with SEO,
·Capture meaningful metric’s from NCSA raw web logs,
·Use “relativistic coding” in time series analysis,
·Aggregate the dynamic quality of streaming data to create new variables
·Quantify text-mining from organic keyword searches on the web,
·Define “high-valued” or “target” segments
·Re-aggregate the hierarchy of data into meaningful segments
·Create dimensions for CRM based upon simple sign-in data
·Look at data and their transformations in new and different way
Featured
Advertisement
Analytic Jobs
Most Visited
© 2010 Created by Vincent Granville
Comment Wall
You need to be a member of AnalyticBridge to add comments!
Join this Ning Network