Statistician/Modeler

Statistician/Modeler
Raleigh, NC

Job Description

As part of the modeling team, you will create the analytics that help drive the business at Andrew Davidson & Co., Inc.

You will develop and improve credit models for mortgages and other asset classes using the most recent methods from both statistics and machine learning. You will not only have a deep understanding of your research, but you will also be able to clearly communicate useful results to both modelers and model users.

As a statistician/modeler and machine learning/data analytics researcher you will:

  • Develop and maintain prepayment and credit models for mortgages and other consumer and commercial loans.
  • Identify, clean and manipulate data to address missing or incorrect information, choose proper statistical models and parameterize the models.
  • Interpret data trends, economic influences, and other factors to improve our modeling efforts.
  • Validate and document the statistical models for internal and external use.
  • Write and present technical and non-technical reports about the models and its applications.
  • Monitor developments to determine the need for new or improved models and analytical procedures.

Qualifications

    1. Fresh PH.D. or Master’s in statistics or in a related field to statistics
    2. Some experience with mortgage credit modeling (prepayments, defaults, severities)
    3. Modeling experience in the origination space
  • You’ve worked with statistical models such as linear and non-linear regression, logistic regression, and generalized linear models.
  • You understand performance measures such as area under the curve (AUC), and the receiver operating characteristic (ROC) curve.
  • You are proficient in packages such as SAS, SQL, and C++.

Good to Have

  • You know machine learning techniques.
  • You have some knowledge about fixed-income finance.

Please send your resume and a cover letter in confidence to careers@ad-co.com with the title of the position that you're interested in the subject line.