The best preparation is a rigorous foundation in calculus, linear algebra, probability, statistics, microeconomics, econometrics, and programming. The exact emphasis changes with the kind of research you may pursue.
The common core
- Calculus: differential and integral calculus; multivariable calculus if your schedule allows.
- Linear algebra: vectors, matrices, systems, eigenvalues, and the geometry behind regression.
- Probability and mathematical statistics: distributions, estimation, hypothesis testing, and asymptotic reasoning—not only a business statistics survey.
- Intermediate microeconomics: optimization, incentives, uncertainty, information, and equilibrium.
- Econometrics: regression, identification, panel data, and causal inference.
- Programming: enough fluency to clean data, reproduce an analysis, debug your work, and document it for someone else.
Published expectations differ. Illinois explicitly names calculus and linear algebra; Stanford names probability and statistics, linear algebra, calculus, optimization, microeconomics, econometrics, and programming; other programs publish no formal list. Silence on an admissions page does not mean the first-year courses will slow down to teach the prerequisites.
If you have limited room
Take the most rigorous version you can complete well, in this order:
- Calculus and linear algebra
- Probability or mathematical statistics
- Intermediate microeconomics
- Econometrics
- A programming or statistical computing course
Real analysis is standard preparation for economics Ph.D. programs, but it is not a published requirement for the accounting doctoral programs reviewed for this site. It can be valuable for analytical research and as evidence of mathematical maturity. For most accounting applicants, however, econometrics, mathematical statistics, linear algebra, and programming have higher immediate value.
Preparation by research method
| Likely method | Prioritize | Useful next courses |
|---|---|---|
| Archival | Econometrics, linear algebra, probability, statistical programming | Panel data, causal inference, databases, Python or R, text analysis |
| Experimental | Probability, statistics, research design, psychology | Experimental design, judgment and decision-making, measurement, nonparametric methods |
| Analytical | Multivariable calculus, linear algebra, probability, microeconomic theory | Real analysis, game theory, optimization, advanced microeconomics |
| Qualitative or field | Research design, theory, measurement, writing | Interview methods, case research, survey design, field methods, mixed methods |
You do not need to choose a method permanently before applying. A working hypothesis simply helps you use scarce electives well.
What software to learn first
Learn one tool deeply enough to complete a small project without a tutorial open at every step.
- R or Stata is a practical first statistical environment. Which one is best depends on the programs and faculty you may work with.
- Python is especially useful for data acquisition, text, automation, and machine-learning work.
- SAS remains important in parts of archival accounting and in legacy code built around WRDS. Ask your target program before investing heavily.
- SQL and version control are quiet force multipliers. They make data work and collaboration safer even when no course requires them.
Do not try to collect four beginner certificates. Depth transfers; a superficial tour of every package does not.
For BYU Prep Track students
Use BYU's official Ph.D. Prep Track page for current requirements, course numbers, electives, outlines, and advising contacts. Those details change; the durable goal is to build enough depth in economics, mathematics, statistics, research design, and programming to begin doctoral coursework well prepared.
If you are not at BYU
You do not need a formal prep track. The relevant courses often live in mathematics, statistics, economics, computer science, and psychology rather than the business school. Pair coursework with research experience: admissions committees care whether you understand what the tools are for, not only whether they appear on a transcript.