Data defines the model by dint of genetic programming, producing the best decile table.
GenIQ Articles: Analytic
Bruce Ratner, Ph.D.


    1. A Data Mining Method for Moderating Outliers, Instead of Discarding Them
    2. The Originative Statistical Regression Models: Are They Too Old and Untenable?
    3. The Predictive Model: Its Reliability and Validity
    4. Life-Time Value Modeling of Big-ticket Items
    5. Validating the Logistic Regression Model: Try Bootstrapping
    6. Regression Modeling Involves Art, Science, and Poetry Too
    7. Re-Data-Mining Your Constantly-updated Database: A Criterion for Doing So
    8. What Criteria Do You Use to Build a Model that Maximizes the Cum Lift?
    9. What Criteria Do You Use to Determine the Best Model?
    10. Top Five Statistical Modeling Problems: Nonissues for the Machine-learning GenIQ Model
    11. Statistical vs. Machine-Learning Data Mining
    12. CHAID-based Data Mining for Paired-Variable Assessment
    13. The Missing Statistic in the Decile Table: The Confidence Interval
    14. The Importance of Straight Data: Simplicity and Desirability for Good Model Building Practice
    15. The Paradox of Overfitting
    16. Building a Database Model to Outperform a Test Campaign
    17. To Fit or Not to Fit Data to a Model
    18. Assessing the Predictiveness of a Classification Model: Traditional vs. Modern Methods 
    19. Two-by-Two Classification and Decile Tables - A Comparison
    20. Genetic vs. Statistic Regression Models - A Comparison
    21. Your Customers are Talking: Are You Listening?
    22. Is Not a Response-Model Tree a Response-Model Tree by Any Other Name?
    23. Interpretation of Coefficient-free Models
    24. Social Network Analysis, Social Media Data, and Text Mining to Boost Business Intelligence
    25. Predictive Modeling Using Real-time Data
    26. Data Mining Quiz - II
    27. Data Mining Quiz
    28. CHAID: Nine Inventive, Utile Applications Beyond Its Original Intent
    29. Response-Approval Model: An Effective Approach for Implementation
    30. Data Mining: Illustration of the Pythagorean Theorem
    31. Stepwise is a Problematic Method for Variable Selection in Regression: Alternative Methods are Available
    32. What If There Were No Significance Testing?
    33. A Simple Method for Assessing Linear Trend and Seasonality Components in Database Models
    34. Variable Selection Methods in Regression: Ignorable Problem, Outing Notable Solution
    35. A New CRM Method for Identifying High-value Responders
    36. CRM Segmentation for Targeted Marketing
    37. Retain Best Customers and Maximize their Potential: A CRM Machine-learning Approach
    38. A New CRM Method for Identifying High-value Responders
    39. Predicting the Quality of Your Statistical Regression Models
    40. Confusion Matrix: Perhaps Confusing, but Definitely Biased
    41. What is the GenIQ Model?
    42. Linear Probability, Logit, and Probit Models: How Do They Differ?
    43. A Database Marketing Regression Model that Maximizes Cum Lift
    44. A New Method of Modeling Missing Data: Deliverance of Discarded, Incomplete Cases
    45. Predicting Share of Wallet without Survey Data
    46. Variable Selection Methods in Regression: Many Statisticians Know Them, But Few Know They Produce Poorly Performing Models
    47. Statistical Modelers and Data Miners: Variable Selection, Data Mining Paradigm, Optimal Decile Table, and more ...
    48. The GenIQ Model: Data-defined, Data Mining, Variable Selection, and Decile Optimization
    49. When Data Are Too Large to Handle in the Memory of Your Computer
    50. How To Bootstrap
    51. Data Mining: An Ill-defined Concept
    52. HELP! I Need Somebody, Not Just Anybody ...
    53. Do-It-Yourself Method for Finding the Square Root of 2 
    54. GenIQ: A Visual Introduction
    55. Overfitting: Old Problem, New Solution
    56. Genetic Data Mining: The Correlation Coefficient
    57. Data Cleaning is Not Completed Until the “Noise” is Eliminated
    58. How to Make the Best Credit Score Even Better
    59. Multivariate Regression Trees: An Alternative Method
    60. "Grand" words (1000) about the GenIQ Model.
    61. Historical Notes on the Two Most Popular Prediction Models, and One Not-yet Popular Model
    62. Real World Data are Dirty: Data Cleaning and the "Noise" Problem
    63. GenIQ: For Modelers Who Daringly Consider a Different Model –
    64. The Most Compelling Illustration of the GenIQ Model
    65. A Most Compelling Illustration of the GenIQ Model
    66. GenIQ Lets the Data Specify the Model
    67. Data Mining Using Genetic Programming
    68. GenIQ-enhanced Regression Model
    69. GenIQ-enhanced/Data-reused Regression
    70. GenIQ: Nonlinear Curve Fitter
    71. GenIQ: OLS Curve Fitter
    72. A Method for Moderating Outliers, Instead of Discarding Them
    73. Building Statistical Regression Models: Straight Data are Necessary
    74. Logistic Regression versus Machine Learning Regression
    75. Ordinary Regression versus Machine Learning Regression
    76. The GenIQ Model: FAQs
    77. Interpreting Model Performance: Use the “Smart” Decile Analysis
    78. Predictor Variable Importance: Multicollinearity is Not a Problem for a Genetic Regression Model
    79. Dummy Variables: The Problem and Its Solution
    80. Finding the Best Variables for Database Marketing Models
    81. Decile Analysis Primer: Cum Lift for Response Model
    82. Maximizing the Lift in Database Marketing
    83. When Statistical Model Performance is Poor: Try Something New, and Try It Again
    84. A Hybrid Statistics-Machine Learning Paradigm for Database Response Modeling
    85. Tukey's Bulging Rule: Why Use It, and What to Do When It Fails
    86. Tukey's Bulging Rule for Straightening Data
    87. Modeling a Skewed Distribution with Many Zero Values
    88. A New Jackknife Method: 3-in-1 Tool for Variable Selection, Data Mining and Model Building
    89. A Genetic Model to Identify Titanic Survivors
    90. Statistics versus Machine Learning: A Significant Difference for Database Response Modeling
    91. The Genetic Programming Engine that Does: Data Specify the Model, Not Fit Data to a Model
    92. GenIQ-Parkinson's Law: The GenIQ Model Expands to Fill the Time Available for Model Completion
    93. Genetic Data Mining Method for the Proper Use of the Correlation Coefficient
    94. A Genetic Imputation Method for Database Modeling
    95. Missing Value Analysis: A Machine-learning Approach
    96. A Genetic Logistic Regression Model: A Model-free Approach to Identifying Responders to a CRM Solicitation
    97. Predictive Analytics Now Accessible to Excel Spreadsheet Users: GenIQ Model Software with an Excel Toolbar
    98. An Alternative Response Model
    99. Analysis and Modeling for Today's Data
    100. Using the GenIQ Model to Insure the Validation of a Model is Unbiased
    101. Gain of a Predictive Information Advantage: Data Mining via Evolution
    102. Response-Approval Model: An Effective Approach for Implementation
    103. Marketing Optimization Model: A Genetic Approach
    104. Binary Logistic Regression: A Model-free Approach
    105. Ordinal Logistic Regression: A Model-free Approach
    106. Multinomial Logistic Regression: A Model-free Approach
    107. Quantile Regression: Model-free Approach
    108. Rethink The Regression Model: Think GenIQ Model
    109. Building a CRM Model for Identifying Profitable Leads: The Genetic Contact-Profit Model
    110. A New Method of Decile Analysis Optimization for Database Models
    111. Multiple Catalog Mail Campaigns: Who Gets Mailed Next, and Which Catalog Should It Be?
    112. Building and Solving Response Optimization Models with the GenIQ Model
    113. Gaining Insights from Your Data: A Neoteric Machine Learning Method
    114. Data Mining for the Desktop
    115. Radically Distinctive Without Equal Predictive Model
    116. Extracting Nonlinear Dependencies: An Easy, Automatic Method
    117. Retail Revenue Optimization: Accounting for Profit-eating Markdowns

For more information about this article, call Bruce Ratner at 516.791.3544 or 1 800 DM STAT-1; or e-mail at br@dmstat1.com.
Sign-up for a free GenIQ webcast: Click here.


DM STAT-1 CONSULTING / br@dmstat1.com
574 Flanders Drive / North Woodmere, NY 11581 / U S A
Voice 1-516-791-3544 / Fax 1-516-791-5075
Toll Free 1 800 DM STAT-1