91ÊÓƵ

The specific objectives of this course are that by the end of the semester you will be able to:

• Write and speak about your statistical work, status progress, and career development aspirations to both technical and non-technical audiences.
• Produce graphics, reports, and deliverables (inclusive of job application deliverables) that communicate clearly and effectively.
• Approach group projects with specific ideas about how to work effectively with others, manage capacity, and divide responsibilities.
• Understand goal setting, delivery leading practices, methods, and common deliverables, client-facing expectations, and the consulting perspective.
• Effectively engage stakeholders.

By the end of this course, students will be able to:

• Practice the Fundamentals of Tidy Data Wrangling and Reproducible Workflows.
  • Distinguish between data types and pinpoint which graphics and analyses are appropriate for a particular data type.
  • Write easily readable and reproducible code to explore datasets graphically.
  • Practice tidy data manipulation in R using the tidyverse with consistent code style.
  • Practice reproducible data analysis workflows with RMarkdown.
• Create High-Quality Statistical Graphics.
  • Master the use of ggplot2 to create statistical graphics that are easily readable and understandable for technical and non-technical audiences.
  • Incorporate statistical information (e.g., the results of statistical tests or uncertainty quantification) into elegant data visualizations.
  • Create interactive visualizations enabling end-users to explore data with dashboards and animations.
• Assess and Critique Statistical Graphics.
  • Review others’ statistical graphics objectively and professionally.
  • Describe the pros and cons of a given graphical choice.
  • Give useful critiques, feedback, and suggestions for improvement on others’ graphics.
• Write 91ÊÓƵ Statistical Graphics and Data Visualizations.
  • Describe graphics concisely and accurately to technical and non-technical audiences.
  • Incorporate appropriate statistical language in written descriptions of graphics.

By the end of this course, students will be able to:

• Design SQL tables to store complex, frequently updated data.

• Write SQL queries to group, aggregate, and summarize data stored in multiple interconnected tables.

• Use Python to execute SQL queries and dynamically act upon stored data.

• Load data into SQL databases and update existing data.

• Develop Python data pipelines that ingest data from multiple sources, load it into a database, and produce automated reports summarizing that data.

• Run data pipelines on cloud computing resources.

• Work with data stored in distributed file systems.

By the end of this course, students will be able to:

• Load large quantities of data into Spark.

• Use Spark to do analytics and reporting on large distributed datasets.

• Use Spark’s machine learning features to fit models on large distributed datasets.

• Fit machine learning models on a cloud platforms such as Azure.

Upon completing this course, students should be able to:

• Understand the machinery of linear regression well enough to use it intelligently.

• Build, fit and critically evaluate linear regression models and some generalizations of them in a variety of messy, real-world settings.

• Communicate statistical results clearly in writing, from sentence to paragraph to full report, especially using the IDMRaD format.

• Carry out the ABA−1 process: translate from real world to quantitative terms, analyze quantitatively, and translate back to real world.

Students will learn to:

• Recognize the goals, constraints, and limitations involved in designing an experiment for a particular problem.
• Design an appropriate experiment to answer specific substantive questions.
• Analyze experimental data using a variety of statistical methods in R.
• Use simulations and power analyses to explore the weaknesses of a given design.
• Communicate design decisions, analyses, and substantive results in written reports.

Upon completing this course, students should be able to:

• Use the command line both locally and on remote machines.
• Develop correct, well-structured and readable code.
• Design useful unit tests at all stages of development.
• Effectively use development tools such as editors/IDEs, debuggers, prolers, testing frameworks and a version control system.
• Select algorithms and data structures to solve common statistical and computing problems.
• Write programs in Python that are well-designed and facilitate code reuse and generalization.
• Analyze computational scaling of programs and use parallelization to speed up run-times.

By the end of the course, students will be able to:

• Explain fundamental topics in sports analytics and the statistical methods used to address problems in this area.
• Build and interpret statistical models for sports data and quantify the associated uncertainty.
• Recognize a sports analytics problem and develop an appropriate modeling approach, including formally specifying models, stating assumptions, and implementing them.
• Critique public sports analytics work from a statistical perspective.
• Use R toolkits to implement relevant techniques for sports analytics problems.
• Communicate the findings from the sports analytics project clearly, using explanations and visuals appropriate for technical and non-technical audiences.
• Develop a sports analytics project suitable for inclusion in a personal portfolio.

Upon completing this course, you should be able to tackle new Statistical Learning problems, by:

• Selecting the appropriate methods and justifying your choices.
• Implementing these methods programmatically (using, say, the R programming language) and evaluating your results.
• Explaining your results to a researcher outside of statistics or computer science.

At the completion of this course, students will:

• Understand the role and significance of statistical methods in modern finance and in the analysis of large and complex financial datasets.
• Explain fundamental finance concepts and communicate financial data and analyses effectively to professional audiences.
• Locate and access standard sources for financial data using Python.
• Analyze the distributional properties of financial returns, and identify their relevance to important financial models.
• Apply basic methods and tools used in time series analysis, and understand their relevance in modelling financial data.
• Implement and interpret factor models, including the CAPM model, in finance.
• Execute data cleaning, manipulation, and analysis in Python, acknowledging the importance of Python in the finance industry.
• Describe the common properties of stochastic processes in finance, including Brownian motion, the Poisson process, and Markov chains.
• Summarize the foundational principles of risk management as practiced in finance. 

By the end of the course, you will be able to:

• Load text in R and clean it, extract common linguistic features, and compute summary
  statistics.
• Interpret linguistic analyses and effectively pair those methods with research questions (i.e., to know when/why a particular method will/will not work based on research goals and/or data type).
• Conduct statistical analyses appropriate for linguistic features, including dimension reduction, keyness comparisons, and multivariate tests.
• Design analyses appropriate for quantitatively based linguistic research.
• Develop a project and report it clearly and rigorously for a variety of audiences.

Upon successful completion of this course, students should be able to select, execute, and interpret analyses of time series data and communicate the results of time series analyses to clients and colleagues.

The main purpose of this course is to help students develop skills in interacting with a client and in digging more deeply into problems that involve statistical practice.

The application cycle for the MADS program is from Monday, October 5, 2026, until Friday, January 15, 2027 at 11:59 pm EST. We will accept applications thereafter on a rolling basis after the regular deadline as long as space remains in the incoming class but recommend applications are submitted prior to the cycle deadline to ensure an application review. The will be open on October 5.

For a completed application submission, applicants must provide each of the following, . Please note that we do not accept mailed applications or mailed supplemental materials. All materials must be uploaded before the application deadline.

• Application fee
• Undergraduate and graduate transcripts of all institutions where you have attended and/or graduated
• Applicant background, goals and post-graduation objectives
• A resume
• Three recommendation letters uploaded by recommenders (two of which should be faculty recommendations)
• If applicable, GRE General Test scores. Our GRE codes are 2074 (Institution) and 0705 (Department).
• If applicable, TOEFL, IELTS test, or Duolingo scores. Our TOEFL codes are 2074 (Institution) and 59 (Department)

Application Fees

If you apply by November 2, 2026, the application fee is $50.00. From November 3, 2026 forward the program application fee is $75.00.

Prerequisite Coursework

At the time of applying to the MADS program, successful applicants are highly encouraged to have:

• Two semesters of calculus-based probability and mathematical statistics (91ÊÓƵ prerequisite classes are 36-225 and 36-226).
  
  Topics should include: random variables, distribution functions, joint and conditional distributions, functions of random variables & their probability distributions; maximum likelihood estimation, properties of estimators, hypothesis testing, interval estimation.
  
  Typical Textbook: Mathematical Statistics with Applications - Wackerly, et al. or equivalent
• One course in linear regression analysis (a 91ÊÓƵ prerequisite course is 36-401)
  
  Topics should include: exploratory data analysis, linear regression models, validation and interpretation of models
  
  Typical Textbook: Applied Linear Regression Models - Kutner, et al. or equivalent. A good Econometrics course is an acceptable substitute.
• Familiarity with/exposure to matrix algebra and/or linear algebra. Applicants should be familiar with vectors and matrices, matrix multiplication, and inverses.

Test Scores: TOEFL, IELTS, and Duolingo

If your native language (language spoken from birth) is not English, we require a current English proficiency score report. The English proficiency requirement cannot be waived for any reason. For all applicants whose native language is not English, we strongly encourage TOEFL or IELTS scores. If you need to retake your TOEFL or IELTS for this application, we highly recommend you take the exam well before the deadline. Reportable scores can take two weeks to process and we may not be able to review any scores that arrive after our deadline. Our TOEFL codes are 2074 (Institution) and 59 (Department). We do not accept expired scores.

However, we understand that in some cases it may not be possible for an applicant to test for TOEFL or IELTS.  If you are not a native English speaker, and despite your best efforts are unable to test for TOEFL or IELTS, we will accept the Duolingo English proficiency test.

Minimum Scores

• TOEFL total score - 5
  • TOEFL Speaking Score - 5.5
• IELTS speaking score - 9
• DuoLingo - 135

Note: Previous studies at a U.S. high school, college, or university are NOT grounds for a language-test waiver.  Additionally, previous studies at a non-U.S. English language instruction high school, college, or university are also NOT grounds for a language-test waiver. 

All scores must be uploaded and received before the deadline in order to assure an equitable review of your application.

The tuition for the MADS program for the 2026-2027 academic year is $59,600 plus mandatory student fees of approximately $1035.

As part of the admissions offer, yet subject to eligibility and qualifications, MADS students will receive a stipend as an educational assistant or teaching assistant in the Department of Statistics & Data Science. Outside of the teaching assistant role, the MADS program does not offer any additional funding opportunities.

Students applying for federal financial aid programs must complete a Free Application for Federal Student Aid (FAFSA) form. Visit to complete the FAFSA form. We recommend that admitted students discuss financial aid options with the university’s central HUB student service center. The HUB website provides the most comprehensive knowledge on the graduate financial process.

Loan Options for U.S. Citizens

For U.S. citizens and permanent residents, there may be eligibility for federal Direct student loans, which can include Unsubsidized and Grad Plus Loans.  The Graduate Financial Aid Process can direct you to understand the financial aid process and the financial aid timelines.

Through , a free loan comparison service,  you can learn more about private (non-federal) education loans and supplemental federal loans.

Loan Options for Non-U.S. Citizens

Private loan options are available to non-U.S. citizens. Generally, private education loans may require non-U.S. citizens to have a U.S. cosigner, yet loan options may be available to qualified non-U.S. citizens without a cosigner.

can provide you with a list of lenders who have issued and disbursed loans to Carnegie Mellon students in the past 5 years.

Who should apply?

Students who are an ideal fit for the MADS program desire a career in a fast-paced, high-growth industry in the United States immediately following graduation. MADS graduates have successfully gained positions as data scientists, data analysts, and data engineers in diverse industries such as banking, sports, health care, government, and tech. 

There are a variety of programs at 91ÊÓƵ for students who are interested in related specific statistical graduate study topics and related outcomes. The Master of Science in Computational Finance (MSCF) is a program for students specifically interested in quantitative finance, and the is a program for students specifically interested in advanced study in machine learning.

Do I need to have a degree in statistics to be considered for MADS?

We welcome all majors yet we ask that students demonstrate academic success in the prerequisite courses in order to maximize success in the program. Please see the prerequisites below.

What are the prerequisites for the program?

The program is highly rigorous and intense. We want our students to be set up for success in the classroom. At the time of applying to the MADS program, successful applicants are highly encouraged to have: 

• Two semesters of calculus-based probability and mathematical statistics (91ÊÓƵ prerequisite classes are 36-225 and 36-226)
  
  Topics should include: random variables, distribution functions, joint and conditional distributions, functions of random variables & their probability distributions; maximum likelihood estimation, properties of estimators, hypothesis testing, interval estimation.
  
  Typical Textbook: Mathematical Statistics with Applications - Wackerly, et al. or equivalent
  91ÊÓƵ classes are: 36-225 and 36-226
• One course in linear regression analysis (a 91ÊÓƵ prerequisite course is 36-401)
  
  Topics should include: exploratory data analysis, linear regression models, validation and interpretation of models
  
  Typical Textbook: Applied Linear Regression Models - Kutner, et al. or equivalent. A good Econometrics course is an acceptable substitute.
• Familiarity/exposure with matrix algebra and/or linear algebra. Applicants should be familiar with vectors and matrices, matrix multiplication, and inverses.

Can I take classes outside of the Department of Statistics & Data Science?

MADS students are typically not permitted to take courses outside of the Department of Statistics & Data Science. Normally, all the MADS students take core requirements together as a cohort. Electives are offered on a rotating basis.

Alternative courses within the Department of Statistics & Data Science may be allowed as a substitute should a comparable course of study already have been successfully completed prior to entry into the MADS program — any substitutions are considered on a case-by-case basis and must be approved by the Department of Statistics & Data Science.

Will you review my application if it is incomplete?

To ensure an equitable process for all applicants, we will review only all fully completed applications submitted by January 15, 2027, and only additional fully completed applications submitted after January 15 if there is space available in the program.

A hallmark of the MADS program is a small cohort size to facilitate a high-touch student experience.

What if I cannot take the TOEFL or IELTS?

We understand that in some cases it may not be possible for an applicant to test for TOEFL or IELTS. If you are not a native English speaker, and despite your best efforts are unable to test for TOEFL or IELTS, we will accept the Duolingo English proficiency test. A score of at least 135 overall is desirable.  We will give consideration to those with the following subscores and above:

• Literacy: 135
• Conversation: 130
• Comprehension: 145
• Production: 113

All scores must be uploaded and received before the deadline in order to assure an equitable review of your application.

How long are TOEFL scores valid?

If you need to retake your TOEFL, IELTS and/or GRE for this application, we highly recommend you take the exam no later than November 28, 2026. Reportable scores can take two weeks to process and we may not be able to review any scores that arrive after our deadline.