Data Mining in Pharmaceutical Marketing in addition to Sales Analysis Contents What is the Data Mining Methodology

Data Mining in Pharmaceutical Marketing in addition to Sales Analysis Contents What is the Data Mining Methodology www.phwiki.com

Data Mining in Pharmaceutical Marketing in addition to Sales Analysis Contents What is the Data Mining Methodology

Kingsley, Barbara, Faculty Advisor has reference to this Academic Journal, PHwiki organized this Journal Data Mining in Pharmaceutical Marketing in addition to Sales Analysis Pavel Brusilovskiy, PhD Merck Contents What is Data Mining Data Mining vs. Statistics: what is the difference Why Data Mining is important tool in pharmaceutical marketing research in addition to sales analysis Case Study What is the Data Mining “The magic phrase to put in every funding proposal you write to NSF, DARPA, NASA, etc” “Data Mining is a process of torturing the data until they confess” “The magic phrase you use to sell your – database software – statistical analysis software – parallel computing hardware – consulting services”

UEI College CA www.phwiki.com

This Particular University is Related to this Particular Journal

Data Mining Data Mining is a cutting edge technology to analyze diverse, multidisciplinary in addition to multidimensional complex data is defined as the non-trivial iterative process of extracting implicit, previously unknown in addition to potentially useful in as long as mation from your data Data mining could identify relationships in your multidimensional in addition to heterogeneous data that cannot be identified in any other way Successful application of state-of-the-art data mining technology to marketing, sales, in addition to outcomes research problems (not to mention drug discovery) is indicative of analytic maturity in addition to the success of a pharmaceutical company Data Mining in addition to Related Fields Visualization Machine Learning Statistics Database Data Mining Is Data Mining extension of Statistics Statistics vs. Data Mining: Concepts

Statistics vs. Data Mining: Regression Modeling What is an unstructured problem What are differences between Data Mining in addition to Statistics Statistical analysis is designed to deal with well structured problems: Results are software in addition to researcher independent Inference reflects statistical hypothesis testing Data mining is designed to deal with unstructured problems Results are software in addition to researcher dependent Inference reflects computational properties of data mining algorithm at h in addition to

Is Data Mining extension of Statistics Data Mining in addition to Statistics: mutual fertilization with convergence Statistical Data Mining (Graduate course, George Mason University) Statistical Data Mining in addition to Knowledge Discovery (Hardcover) by Hamparsum Bozdogan (Editor) An overview of Bayesian in addition to frequentist issues that arise in multivariate statistical modeling involving data mining Data Mining with Stepwise Regression (Dean Foster, Wharton School) use interactions to capture non-linearities use Bonferroni adjustment to pick variables to include use the s in addition to wich estimator to get robust st in addition to ard errors When data mining technology is appropriate Data mining technology is appropriate if: The business problem is unstructured Accurate prediction is more important than the explanation The data include the mixture of interval, nominal, ordinal, count, in addition to text variables, in addition to the role in addition to the number of non-numeric variables are essential Among those variables there are a lot of irrelevant in addition to redundant attributes The relationship among variables could be non-linear with uncharacterizable nonlinearities The data are highly heterogeneous with a large percentage of outliers, leverage points, in addition to missing values The sample size is relatively large Important marketing, sales, in addition to outcomes research studies have the majority of these features Accurate prediction is more important than the explanation

Case Study: Effectiveness Evaluation of Vaccine Sales Force New lunched vaccine (10 months in marketplace) Sales as long as ce structure: two sales team Team-1 Team-2 Some locations are visited only by Team-1 (1-up promotion), others – by both teams (2-up promotion) Promotion is on doctors/HCP level, but sales is on location level Business question: What is the effectiveness of 2-up promotion Dependent Variables in addition to Study Design Dependent variables (criteria to judge): Total dosage purchased (shipped or ordered) Total dosage purchased per promotion dollar Probability of making a purchase Time to the first purchase Frequency of purchase, etc. Study Design: Test (2-up promotion locations) – Control (1-up promotion locations) Consider Vaccine sales as two criteria problem: Estimate outcome as long as Test in addition to Control groups, taking into account difference in sales difference in promotion cost Use pre-period data to match Test in addition to Control groups in addition to post-period data to compare sales in addition to promotion cost Independent Variables 47 input variables Demographics of a location Geography, specialty, potentials, potential decile, average age, reimbursement, believes, etc. Promotion activities Number of direct interactions with HCP by Team-1 Number of direct interactions with HCP by Team-2 Percentage of direct interaction with decision maker by Team-1 Percentage of direct interaction with decision maker by Team-2 Number of phone interactions with HCP by Team-1 Number of phone interactions with HCP by Team-2, etc.

Variables Distribution Private Dosage Potential Public dosage Potential Number of Sales Calls Vaccine Dosage Sales Methodology Form Test – Control groups, using only pre-period data in addition to propensity score methodology with greedy one-to-one matching technique on propensity score Develop models as long as the post-period data as long as total vaccine sales, controlling as long as “location demographics” variables promotion variables in pre-period sales in pre-period Estimate the difference in sales as long as Test in addition to Control groups, taking into account promotion cost Propensity Score Propensity score is the predicted probability of receiving the treatment (probability of belonging to a test group) is a function of several differently scaled covariates Propensity-Score = f (location demographics variables, promotion variables, sales variables ), 0 < f < 1 where f is a non-parametric non-linear multivariate function A sample matched on propensity score will be similar across all covariates used to calculate propensity score Propensity score with SAS EM Neural Net was the best modeling paradigm Finding in addition to Business Implication There is no algorithmically significant difference in sales between Test in addition to Control groups, but promotion cost as long as Control group is two times lower than as long as Test group. In other words, 2-up structure does not produce desired/expected outcome as long as Vaccine per as long as mance Phone call / Sales call response curve as long as Vaccine sales, constructed by TreeNet Number of Purchases Number of Purchases Number of Purchases has a strong diminishing returns effect when the number of sale calls becomes greater than thirty five in addition to the number of Phone calls becomes greater than 2 Number of Phone calls Number of Sales calls Phone call in addition to Sales call response surface as long as Vaccine sales, , constructed by TreeNet Sales call is the most effective when a location gets two Phone calls Number of Purchase has a strong diminishing returns effect when the number of Sales calls becomes greater than thirty five in addition to the number of Phone calls becomes greater than 2 Number of Phone calls Number of Sales calls Reference David J. H in addition to , Data Mining: Statistics in addition to More The American Statistician, May 1998, Vol. 52 No. 2 http://www.amstat.org/publications/tas/h in addition to .pdf Friedman, J.H. 1997. Data Mining in addition to Statistics. What’s connection Proceedings of the 29th Symposium on the Interface: Computing Science in addition to Statistics, May 1997, Houston, Texas Padhraic Smyth (2000), An Introduction to Data Mining, Elumetric.com Inc Doug Wielenga (2007), Identifying in addition to Overcoming Common Data Mining Mistakes, SAS Global Forum Paper 073-2007 Kingsley, Barbara Daily Forty-Niner Faculty Advisor www.phwiki.com

Kingsley, Barbara Faculty Advisor

Kingsley, Barbara is from United States and they belong to Daily Forty-Niner and they are from  Long Beach, United States got related to this Particular Journal. and Kingsley, Barbara deal with the subjects like Higher Education; Local News; Student/Alumni Interest

Journal Ratings by UEI College

This Particular Journal got reviewed and rated by UEI College and short form of this particular Institution is CA and gave this Journal an Excellent Rating.