Erich Eduardo Kohn Marchena
Download
Recommend
Messages
Share
Erich Eduardo Kohn Marchena

Erich Eduardo Kohn Marchena

About

Detail

Data Scientist
Santo Domingo, Distrito Nacional, Dominican Republic

Contact Erich regarding: 

work
Full-time jobs
Starting at USD3.5K/month
Flexible work
Starting at USD35/hour
connect_without_contact
Finding mentors

Timeline

work
Job
school
Education
folder
Project

Résumé

Jobs verified_user 0% verified
  • B
    Senior Manager Risk Analytics
    Banco BHD
    Jun 2022 - Mar 2024 (1 year 10 months)
    Credit life cyclefor designing behavioral policies. • Developed decision trees to classify customers likely to default on loans resulting in loans defaults dropping by 40% within 12 months of implementation, using Python and Oracle. • Enhanced credit card limit campaign using decision trees to classify customers likely to use more credit, resulting in increased revenue by 23% on selected customers. • Maintained specialized data repositories for monitoring a range of customer behaviors and experimental treatments. • Proposed and oversaw governance policy on internal programming procedures, resulting in department-wide adoption of on-premise GitHub (Azure) for backing up campaign and modeling processes. • Improved time-to-develop for Cre
  • S
    Senior Product Owner Supporting Advanced Analytics and Data Governance.
    Jun 2022 - Dec 2022 (7 months)
    • Delivered a 20% lead retention boost resulting from implementing business retention strategies and next-best-action based on predictions from LightGBM churn classification model, leading to a US$12 million revenue lift within 11 months. • Designed pipelines for assigning next-best-action for retention leads selected by LightGBM churn classification model improving delivery of leads and automating business logic. • Maintained project backlog for 15 bi-weekly sprints for analytics-based Scrum team dynamic using Azure DevOps.
  • G
    Data Mining Manager
    Grupo Universal
    Jan 2017 - May 2022 (5 years 5 months)
    and actuarial strategies with 3 reports. • Developed Value-at-Risk model using SAS; after first year, portfolio grew 9.2% while risk fell 7.4%. • Improved existing customer segmentation strategy leading to 26% reduced customer attrition in 2021. • Led creation of 15+ Power BI dashboards with live connection to databases serving live content to executives. Supporting commercial and actuarial strategies. • Collaborated in pricing analysis of auto insurance; resulted in improved policy and expanded portfolio by 10%. • Responsible for overseeing and implementing the factor commission model for business executives on a monthly basis. • Streamlined data integration for complete business activity reporting, reducing time from 45 min to 2 min
  • D
    Data Mining Project Lead
    Jan 2017 - May 2020 (3 years 5 months)
  • h
    Statistical Consultant
    high partner sas
    Jan 2016 - Nov 2016 (11 months)
Education verified_user 0% verified
  • SAS Institute Inc.
    SAS Certified Base Programmer for SAS 9
    SAS Institute Inc.
    Jan 2016 - May 2016 (5 months)
  • S
    Certified Predictive Modeler Using SAS Enterprise Miner 14
    SAS
  • Deeplearning.ai
    TensorFlow Developer Professional Certificate
    Deeplearning.ai
  • TOEFL
    iBT: 106/120
    TOEFL
  • Deeplearning.ai
    Mathematics for Machine Learning and Data Science Specialization
    Deeplearning.ai
  • S
    SAS Certified Statistical Business Analyst Using SAS 9
  • C
    Credit Risk Modeling Using SAS
  • P
    Bachelor of Arts
    Pontificia Universidad Cat ´ olica Madre y Maestra
  • PCA Pharmacy
    Certified Associate in Python Programming
    PCA Pharmacy
  • S
    Certified Advanced Programmer for SAS 9
    SAS
Projects verified_user 0% verified
  • C
    Credit Risk Default for Digital Revolving Loans
    Jan 2023 - Jun 2023 (6 months)
    Designed Management Information Systems (MIS) to track and monitor performance for revolving personal loans for digital-only product. Calculated how many months on books (MOB) and delinquency status (0, 30, 60, 90) each client was performing using Oracle. Then, modeled the relationship between 90-day delinquency and other drivers for loans in a Decision Tree using Python. Using the Decision Tree, predicted the behavior of newer loans and designed a credit life-cycle policy with the intent to optimize portfolio returns. Impact: Monthly portfolio performance reporting and portfolio inspection on a quarterly basis; portfolio delinquency dropped from 1.5% to 0.9% within 12 months.
  • I
    Insurance Pricing Analysis
    Apr 2017 - Nov 2017 (8 months)
    Pricing Auto Insurance, Policy In collaboration with actuary department, analyzed pricing structure of automobile insurance and tested business rules for commercial premiums. Discovered discrepancy between amounts of discounts being offered to all clients, which resulted designing policy for normalizing premiums and discount approvals. Designed simulation system for policy changes using what-if analysis using SAS. Impact: Created a Auto-Insurance Simulation Tool in SAS and designed cleaner policy structure for auto-insurance premiums.