Finance · Updated 16 September 2026
Quantitative Analyst (Quant Researcher / Quant Trader) salary and career roadmap
Find statistical edge in market data and turn it into code that trades money - median total comp around $220k across the market, $600k at Citadel.
Salary data checked · outlook from BLS projections released · by Bilal Tahir
- What it is
- A Quantitative Analyst is a person who finds statistical edge in market data and turns it into code that trades money, building and testing predictive signals, pricing derivatives or owning live risk at a hedge fund, a proprietary trading firm or a bank.
- Salary
- A Quantitative Analyst in the United States earns a median of $225,000 entering the field, $400,000 at mid-career and $600,000 at senior level; the top end is $1,500,000. (Levels.fyi quantitative researcher packages by company, widened with BLS Occupational Outlook Handbook figures for mathematicians and financial risk specialists, )
- Outlook
- Employment is projected to grow 10% over the ten years to 2036, with about 2,000 US openings a year. (BLS Occupational Outlook Handbook, 15-2021 Mathematicians and 15-2041 Statisticians, used as a proxy because no BLS occupation code isolates quantitative analysts (13-2054 Financial Risk Specialists covers bank-side risk quants), )
- Time to first job
- A career changer starting from zero typically needs 24 to 48 months at 15-20 hours a week to reach a first offer.
- Cost to get in
- The cheapest verified route in (PhD in a quantitative field, recruited directly) costs about $0; the most expensive costs about $90,000.
- Roadmap
- Our Quantitative Analyst roadmap is 8 steps and about 1,380 study hours; the fastest route in is Prop trading firm graduate programme at about 12 months.
- Degree
- No degree is legally required, but a Quantitative Analyst is one of the few careers on this site where an advanced degree functions as a genuine filter. The US Bureau of Labor Statistics reports that mathematicians and statisticians typically need at least a master's degree, and most hedge fund and proprietary trading research hires hold a PhD or a master's in mathematics, statistics, physics, computer science or financial engineering.
- Automation exposure
- Low. Machine learning tooling raises a Quantitative Analyst's output rather than replacing the judgement about which backtest to believe, so direct automation risk is low - quants are the people who automate other jobs. The real pressure is competitive: the same tools are available to every rival firm and shorten the half-life of any given signal, so more of the work is finding new edge and less of it is harvesting old edge.
What does a Quantitative Analyst do?
A Quantitative Analyst is a person who finds statistical edge in market data and turns it into code that trades money, building and testing predictive signals, pricing derivatives or owning live risk at a hedge fund, a proprietary trading firm or a bank. "Quant" covers three distinct jobs that pay very differently. The same work is also posted as Quantitative Researcher, Quant Trader, Quantitative Developer, Quantitative Strategist (Strat), Financial Engineer.
"Quant" covers three distinct jobs that pay very differently. A quantitative researcher builds and tests predictive signals - cleaning data, engineering features, fitting models, and defending a backtest against the many ways it can lie. A quantitative trader owns live risk: sizing, execution, hedging and the profit and loss of a book, usually on a systematic or semi-systematic strategy. A quantitative analyst or 'strat' at a bank prices derivatives, builds risk models and validates them for regulators. The first two sit at hedge funds and proprietary trading firms (Citadel, Jane Street, Two Sigma, Jump, IMC, Optiver, DE Shaw, WorldQuant); the third sits at Goldman Sachs, JPMorgan, Morgan Stanley and their peers, and pays roughly a third to a half as much.
Who thrives: people with genuine mathematical maturity - probability, linear algebra, statistics, stochastic calculus for derivatives roles - who also write production-quality Python and usually C++, and who are temperamentally sceptical. The defining professional skill is not building models but refusing to believe your own backtest. Most candidates hold a PhD or a master's in a quantitative field (maths, physics, statistics, CS, electrical engineering, financial engineering); prop shops such as Jane Street and Optiver hire strong undergraduates directly and train them.
The honest tradeoffs. Unlike investment banking, this field is close to a meritocracy of demonstrable ability: firms give you timed probability and coding tests and a research take-home, and your school matters less than whether you pass them. But the mathematical bar is real and cannot be shortcut - an adult switching in from an unrelated job realistically needs 18-36 months of serious maths, statistics and programming work, and usually a master's degree, before interviews become winnable. Career risk is also front-loaded onto results: quant traders and researchers with a poor year or two are cut, and the compensation figures you read are medians of survivors. Base salary is a small share of total comp at funds; the bonus is discretionary and performance-linked, so a $600k median can be a $250k year.
Why Quantitative Analyst pay is high
A quantitative fund's profit is close to linear in the quality of its signals and inversely related to the number of people who share them. If a researcher finds a signal that adds even 0.3 Sharpe to a book running $2bn, the incremental profit is tens of millions of dollars a year, and the firm keeps a performance fee of 20-30% on it or, at a prop shop, all of it. The marginal value of one good researcher is therefore enormous and the supply is genuinely scarce: the intersection of research-grade statistics, low-latency engineering and market intuition is small, and the firms competing for it - Citadel, Jane Street, Two Sigma, Jump, Optiver, IMC, DE Shaw - are also competing with big-tech machine learning teams. Pay is concentrated in bonus rather than salary precisely because the value created is measurable and volatile.
What's good
- Compensation at the top of the finance market without the client-service grind: a Citadel quant researcher median around $600k, Two Sigma around $420k.
- Hiring is closer to a meritocracy than anywhere else in finance - timed tests and a research take-home matter far more than which university you attended.
- Hours are humane by finance standards, typically 45-60 a week.
- Genuinely interesting technical work with immediate, unambiguous feedback on whether you were right.
- Skills transfer directly to machine learning roles in technology, which are a real fallback at comparable pay.
- Low automation risk - AI tools amplify a quant's output rather than replacing the judgment about what to trust.
What's hard
- The mathematical entry bar is genuinely high and cannot be shortcut; most people need a master's or PhD, and 18-36 months of preparation from an unrelated background.
- Pay is dominated by discretionary bonus, so a $600k median can be a $250k year after a bad run.
- Performance culture is unforgiving - traders and researchers with two poor years are commonly cut, and published medians reflect survivors.
- The vast majority of your research fails, and you have to keep working anyway.
- Signal half-lives are shortening as more capital and more machine learning chase the same inefficiencies.
- Bank quant roles pay a fraction of fund roles for similar technical work (Goldman Sachs quant researcher median around $150k on levels.fyi versus $600k at Citadel).
What a Quantitative Analyst does all day
- 8:00am: check overnight performance of the live strategies you own; one signal has decayed and you need to know whether it is noise or regime change.
- 8:30am: morning research meeting - five minutes each on what people are working on, with blunt questions about methodology.
- 9:00am-12:00pm: data work. Clean a new alternative dataset, align timestamps to avoid look-ahead, handle survivorship in the security master. Most of research is this.
- 12:00-1:00pm: lunch, usually at the desk or with the pod; on a trading desk, lunch is when the market is quiet.
- 1:00-4:00pm: fit and validate a candidate signal with purged, embargoed cross-validation; apply a multiple-testing correction and watch most of the apparent edge disappear.
- 4:00-5:00pm: write up the result - including the negative ones - and circulate it for review; defend the method to a senior researcher who will try to break it.
- 5:00-6:30pm: pair with an engineer to move an accepted signal into the production pipeline and add monitoring for it.
- Trading roles differ: market hours are spent on risk, quoting and execution; research happens before the open and after the close.
- Typical hours are 45-60 a week at most funds - much lighter than banking - but the intensity while working is higher and the results are measured.
- Roughly 90-95% of research ideas fail out-of-sample. Getting comfortable with that is the job.
Quantitative Analyst salary in 2026: by level
US, annual, USD. Base plus typical bonus where the source reports it.
A US Quantitative Analyst earns a median of $225,000 entering the field, $400,000 at two to four years and $600,000 at senior level, with the top end at $1,500,000. These are total cash figures in US dollars as of September 2026, synthesised from Levels.fyi quantitative researcher packages by company, widened with BLS Occupational Outlook Handbook figures for mathematicians and financial risk specialists. Pay varies about 10-25% by metro.
The single largest driver of pay is firm type, not seniority. Levels.fyi puts the US median for Quantitative Researcher at about $200k-$220k across all firms, but company medians diverge wildly: Citadel about $600k (L1 $386k to L3 $655k, top reported package $1.5m), Two Sigma about $420k, Google about $325k, IMC about $287.5k, JPMorgan about $217.5k and Goldman Sachs about $150.75k. Jane Street pays interns the annualised equivalent of about $300k. First-year total compensation at top funds and prop shops is commonly reported in the $250k-$450k band; Citadel graduate quant researchers in Chicago are reported at $275k-$475k. By the five-year mark at Jane Street, Jump or Citadel, reported comp for the median survivor runs $800k-$1.2m. Bank strat roles are much lower: BLS gives a $117,330 May 2025 median for financial risk specialists (SOC 13-2054) and $126,710 for mathematicians. Base salary at funds is typically $150k-$300k with the remainder discretionary bonus; guarantees are common only in year one.
No BLS code isolates quant researchers and traders. The closest proxies: mathematicians and statisticians, projected to grow 10% from 2025 to 2035 (statisticians 11%) with roughly 2,000 openings a year on a base of 33,500 jobs, and financial risk specialists within the financial analyst family, where BLS reports a May 2025 median of $117,330. The 2,000 openings figure covers the whole mathematician and statistician occupation, not quant seats specifically, which number in the low thousands a year in the US. Demand is driven by the continued shift of trading volume to systematic strategies, the growth of multi-manager platforms and the migration of machine learning into forecasting. Automation risk is low in an unusual way: quants automate other people's jobs, and AI tooling raises their productivity rather than replacing them - but it also compresses the half-life of any given signal, so the work becomes more competitive rather than less.
“Overall employment of mathematicians and statisticians is projected to grow 10 percent from 2025 to 2035, much faster than the average for all occupations.”
Quantitative Analyst salary by city
National bands scaled by metro wage differentials from the BLS May 2025 OEWS release.
Quantitative Analyst pay is highest in San Jose / Silicon Valley (mid-career median about $568,000, ×1.42 the national figure) and lowest among large metros in Salt Lake City (about $380,000). The multiplier moves the offer, not what you keep after rent and state tax.
| Metro | Entry | Mid | Senior | vs national |
|---|---|---|---|---|
| San Jose / Silicon Valley | $319,500 | $568,000 | $852,000 | ×1.42 |
| San Francisco Bay Area | $303,750 | $540,000 | $810,000 | ×1.35 |
| New York City | $288,000 | $512,000 | $768,000 | ×1.28 |
| Seattle | $270,000 | $480,000 | $720,000 | ×1.2 |
| Boston | $258,750 | $460,000 | $690,000 | ×1.15 |
| Washington DC metro | $252,000 | $448,000 | $672,000 | ×1.12 |
| Los Angeles | $243,000 | $432,000 | $648,000 | ×1.08 |
| Chicago | $236,250 | $420,000 | $630,000 | ×1.05 |
| Austin | $236,250 | $420,000 | $630,000 | ×1.05 |
| San Diego | $234,000 | $416,000 | $624,000 | ×1.04 |
| Denver | $229,500 | $408,000 | $612,000 | ×1.02 |
| Philadelphia | $229,500 | $408,000 | $612,000 | ×1.02 |
| Dallas | $225,000 | $400,000 | $600,000 | ×1.0 |
| Minneapolis | $225,000 | $400,000 | $600,000 | ×1.0 |
| Raleigh-Durham | $225,000 | $400,000 | $600,000 | ×1.0 |
| Houston | $222,750 | $396,000 | $594,000 | ×0.99 |
| Atlanta | $220,500 | $392,000 | $588,000 | ×0.98 |
| Phoenix | $213,750 | $380,000 | $570,000 | ×0.95 |
| Miami | $213,750 | $380,000 | $570,000 | ×0.95 |
| Salt Lake City | $213,750 | $380,000 | $570,000 | ×0.95 |
A multiplier raises the number on the offer letter, not what you keep. San Francisco pays about 35% more than the national median for these roles, but median Bay Area rent and California state income tax eat most of that for anyone below the senior rung. Texas, Florida, Washington, Tennessee and Nevada levy no state income tax, which is worth roughly 4-10% of take-home versus California or New York City, where city tax stacks on top of state tax. Run the comparison on after-tax income minus housing before you move. Fully remote roles are the edge case worth chasing: a national pay band spent in a 0.88 cost market beats a 1.35 salary spent in a 1.6 cost market for most people.
How the multipliers are derived
Anchor: the BLS May 2025 OEWS national mean wage across all occupations is $33.54/hr ($69,770/yr). Metro all-occupation means from the same release: San Jose-Sunnyvale-Santa Clara $57.32 (1.71x national), San Francisco-Oakland-Fremont $48.19 (1.44x), Washington-Arlington-Alexandria $44.20 (1.32x), Seattle-Tacoma-Bellevue $44.13 (1.32x), Boston-Cambridge-Newton $43.09 (1.28x), New York-Newark-Jersey City $41.50 (1.24x), Denver-Aurora-Centennial $39.28 (1.17x), Atlanta-Sandy Springs-Roswell $34.57 (1.03x), Chicago-Naperville-Elgin $34.42 (1.03x, May 2024), Dallas-Fort Worth-Arlington $33.96 (1.01x), Phoenix-Mesa-Chandler $33.48 (1.00x). We damp the top end of those raw ratios. All-occupation means exaggerate the gap for the careers on this site, because national pay bands, remote hiring and company-wide equity grids compress geographic spread for high-skill professional roles more than they do for service and hourly work. Levels.fyi shows the same damping: its Bay Area software engineer average total compensation of about $291k sits roughly 1.3x-1.4x the US median, not 1.7x. Non-US multipliers convert local market rates to USD and are directional, not survey-grade.
How to become a Quantitative Analyst
Every route we could verify, with honest time, cost and difficulty.
There are 6 routes we could verify into Quantitative Analyst work. The fastest is Prop trading firm graduate programme at about 12 months; the cheapest is PhD in a quantitative field, recruited directly at about $0. For an adult already working in software or data science, the lateral move produces the most hires: take a quantitative developer, execution engineering or research-platform seat at a fund or a bank on engineering strength, then convert to research internally over 18 to 36 months. For everyone else the realistic answer is a taught master's in financial engineering with published placement data, because funds rarely read a Quantitative Analyst CV that carries no quantitative credential.
PhD in a quantitative field, recruited directly
The most common route into hedge fund and prop-shop research. A PhD in maths, statistics, physics, CS, electrical engineering or operations research, with real research output and strong programming, is the standard profile; firms recruit from PhD programmes and from postdocs. No finance background is required - most firms explicitly prefer to teach markets to a strong scientist rather than teach science to a financier.
Master's in Financial Engineering / Computational Finance (MFE)
A one to two year taught master's (Baruch, Carnegie Mellon MSCF, Princeton MFin, Berkeley MFE, Columbia MAFN, NYU Courant, Chicago Financial Mathematics) is the standard conversion degree for career changers. Programmes publish placement statistics and run their own recruiting pipelines; the QuantNet rankings are the usual comparison point. Cost is typically $60k-$120k in tuition plus forgone income.
Prop trading firm graduate programme
Jane Street, Optiver, IMC, SIG and Jump hire strong undergraduates and master's students into trader and researcher programmes on the strength of timed probability, mental-maths and game tests, then train them. School matters less than test performance, but the tests are hard and the classes are tiny.
Software engineer or data scientist to quant (internal or lateral)
The most accessible route for an adult already working in tech. Move into a quantitative developer, execution engineering or data-platform role at a fund or bank - these hire on engineering strength - then convert to research internally over 18-36 months by contributing to signal work. Slower, but you are paid while you learn and the credential risk is low.
Bank strat / model validation / market risk, then move to the buy side
Banks hire quantitative analysts for derivatives pricing, XVA, model validation and market risk with a lower bar than funds, often from master's programmes. Pay is roughly $130k-$250k, but the seat is legitimate quant work and a common springboard to a fund after two to four years. FRM or CFA can help here in a way they do not at funds.
Self-taught with a public research track record
Rare but real, mostly at smaller funds and crypto firms: build and publish genuinely rigorous research - a QuantConnect or Kaggle track record, open-source contributions to backtesting libraries, or papers - and apply directly. Works only if the mathematics is real; a strategy backtest with no out-of-sample discipline actively hurts you.
Quantitative Analyst roadmap: 8 steps, 1,380 hours
Becoming a Quantitative Analyst from zero takes about 1,380 study hours across 8 steps, roughly 24 to 48 months at 15-20 hours a week plus a job search. Step one is Fix the mathematics floor first - honestly assess it, then close the gap. The mathematics floor comes first because quant interviews are mathematics interviews with a market veneer, and because it is the cheapest available test of whether the switch is realistic for you - you learn the answer in month two rather than month twenty. Real research with purged, embargoed cross-validation sits in the middle because it is the only portfolio a self-taught candidate can build, and interview drilling comes last because speed and calm decay if you practise them a year early.
0 of 8 steps done · saved in this browser ·
-
1
Multivariable calculus, linear algebra (eigen-decomposition, SVD, projections), and calculus-based probability up to conditional expectation and the central limit theorem. If you cannot compute E[X | Y] for a joint density or explain what an eigenvector of a covariance matrix means, start here and do not skip ahead.
Why now: Quant interviews are mathematics interviews with a market veneer. No amount of finance reading substitutes for the ability to manipulate probability fluently under time pressure, and every downstream topic - stochastic calculus, factor models, machine learning - assumes this material cold. This is also the step where most career changers discover whether the transition is realistic for them, which is worth learning in month two rather than month twenty.
- course Mathematics for Machine Learning Specialization (Linear Algebra, Multivariate Calculus, PCA)Coursera (Imperial College London) · 70 h · Included in Coursera Plus ($59/mo or $399/yr); free to audit
- course 18.S096 Topics in Mathematics with Applications in FinanceMIT OpenCourseWare · 60 h · Free (Creative Commons)
- book Introduction to Probability (2nd Edition)Blitzstein & Hwang (CRC Press) · 120 h · Free lectures and problem sets online; book $70-$95
-
2
Python to a professional standard: NumPy and pandas fluency, vectorisation, profiling, testing with pytest, version control, and writing code another person can run. Then add either C++ (for latency-sensitive trading roles) or heavy data engineering (SQL plus a columnar store such as kdb+ or ClickHouse for tick data).
Why now: Research at a fund is shipped as code that runs on live data, and every interview includes a coding screen. Candidates who can derive the maths but write slow, untested, unreadable Python get filtered out at the first technical round. Firms also test data handling explicitly, because most research time goes on data plumbing rather than modelling.
- course Python for Everybody Specialization (5 courses)Coursera (University of Michigan) · 80 h · Included in Coursera Plus ($59/mo or $399/yr)
- practice Introduction to Financial Python (14-article series)QuantConnect · 30 h · Free
- practice LeetCode - algorithms and data structures practice for quant coding screensLeetCode · 110 h · Free tier; Premium $35/month or $159/year
-
3
Regression and regularisation, bias-variance, resampling, hypothesis testing and multiple-comparison correction, time-series methods (stationarity, ARMA, GARCH, cointegration), and supervised learning with gradient boosting and neural networks. Critically: cross-validation that respects the arrow of time, and purged/embargoed splits for financial data.
Why now: Financial data has a signal-to-noise ratio far worse than anything in a standard ML course, is non-stationary, and is heavily autocorrelated. Standard k-fold cross-validation leaks the future into the past and produces backtests that look wonderful and lose money. Firms test exactly this distinction, because it is what separates someone who has done research on markets from someone who has done Kaggle.
- course Machine Learning Specialization (3 courses, Andrew Ng)Coursera (Stanford Online / DeepLearning.AI) · 95 h · $49/month, or included in Coursera Plus; free to audit
- book Advances in Financial Machine LearningMarcos Lopez de Prado (Wiley) · 60 h · $60-$85
- book An Introduction to Statistical Learning (with Applications in Python)James, Witten, Hastie, Tibshirani & Taylor (Springer) · 45 h · Free PDF; print edition $50-$80
-
4
How exchanges and order books work, what market makers do, bid-ask spread and adverse selection, futures and options mechanics, the volatility surface, and the arithmetic of transaction costs and market impact. For derivatives and bank roles, add stochastic calculus and Black-Scholes derivation, Greeks and hedging.
Why now: A model that ignores the spread, the fee schedule and the impact of your own order is not a strategy. Traders in particular are hired on market intuition - why a spread widens, what happens to implied volatility into an earnings print - and researchers are expected to know what is mechanically possible to trade. This is also where derivatives-heavy bank strat roles live.
- course Financial Engineering and Risk Management Specialization (5 courses)Coursera (Columbia University) · 50 h · Included in Coursera Plus ($59/mo or $399/yr)
- book Options, Futures, and Other Derivatives (11th Edition)John C. Hull (Pearson) · 70 h · $120-$260 new; earlier editions are far cheaper and adequate
- book Trading and Exchanges: Market Microstructure for PractitionersLarry Harris (Oxford University Press) · 30 h · $65-$90
-
5
Take three ideas - a cross-sectional equity factor, a time-series momentum or mean-reversion strategy, and an options volatility strategy - and research each properly: data cleaning, feature construction, out-of-sample testing with purged splits, realistic transaction costs, capacity analysis, and a written record of what you tried and what failed. Deploy at least one in paper trading for three months.
Why now: This is the portfolio. Interviews at funds increasingly include a research take-home, and the judgement being assessed is whether you can distinguish a signal from a story. Paper trading for a quarter teaches you the difference between backtested and live performance in a way no book does. It is also the only credential available to a self-taught candidate.
- practice QuantConnect research and backtesting platform (free tier: unlimited backtesting, all asset classes, 500MB workspace)QuantConnect · 200 h · Free tier; paid Researcher/Team tiers for live trading and more compute
- course CS 7646 Machine Learning for Trading (lecture materials free via Udacity/Ed)Georgia Institute of Technology · 60 h · Free lecture materials; paid as an OMSCS course
- course Machine Learning for Trading Specialization (3 courses)Coursera (Google Cloud & New York Institute of Finance) · 25 h · Included in Coursera Plus ($59/mo or $399/yr)
-
6
If you are under about 30, have strong maths and can fund it, a top MFE (Baruch, Carnegie Mellon MSCF, Princeton, Berkeley, Columbia, NYU Courant) is the fastest reliable conversion, at $60k-$120k tuition and 12-24 months, with published placement data and firm relationships. If you already work in software or data science, the cheaper route is lateral: take a quant developer, execution engineering or data role at a fund or bank and convert internally. A PhD is the right answer only if you genuinely want to do research for five years.
Why now: Funds screen hard on quantitative credentials because the research take-home and interview process is expensive to run. A degree from a programme with a recruiting pipeline gets your CV read; without one, you need either an existing quant-adjacent job or an exceptional public track record. Compare programmes on published employment-at-three-months and salary data, not on brand.
- practice QuantNet MFE programme rankings and placement statisticsQuantNet · 10 h · Free
- cert Financial Risk Manager (FRM) Parts I and II - the cheapest credible signal for bank risk and strat rolesGARP · 480 h · $400 enrollment + $600 early/$800 standard per part
- practice QuantNet and Wall Street Oasis quant forums (programme reviews, offer data)QuantNet / Wall Street Oasis · 20 h · Free
-
7
Timed probability and expectation puzzles, martingale and optional-stopping problems, market-making games (quote a two-sided price, manage inventory), mental arithmetic drills, algorithmic coding rounds, and a research discussion of your own projects. Work the standard books cover to cover, then practise aloud with a timer.
Why now: Prop shops and funds screen almost entirely on timed tests - Optiver and Jane Street style mental-maths and market-making rounds decide whether you progress before anyone reads your CV. The question sets are well-known and finite; the differentiator is speed and calm, both of which come only from volume. Budget 100+ hours and do not attempt interviews before you have.
- book A Practical Guide To Quantitative Finance Interviews (the 'Green Book')Xinfeng Zhou · 50 h · $30-$50
- book Quant Job Interview Questions and Answers (2nd Edition)Mark Joshi, Nick Denson & Andrew Downes · 40 h · $40-$60
- practice Timed mental-maths and market-making practice (Optiver-style 80-in-8 drills)Optiver · 30 h · Free
-
8
Split your list three ways: prop shops and top funds (Jane Street, Citadel, Two Sigma, Jump, Optiver, IMC, DE Shaw) where the bar is highest and comp reaches $400k+ in year one; smaller systematic funds and multi-manager pods where the bar is lower; and bank strat, model validation and market risk desks, which hire more people at $130k-$250k and serve as a springboard. Apply to all three tiers simultaneously.
Why now: The levels.fyi data shows how wide the spread is: a median Citadel quant researcher package is around $600k while Goldman Sachs sits near $150k for the same title. Starting at a bank or a smaller fund is not a dead end - two to four years of genuine research or strat work is the most common way career changers reach a top fund. Applying only to the top tier usually means applying to nothing.
- practice Levels.fyi quantitative researcher compensation by companyLevels.fyi · 5 h · Free
- practice QuantNet job board and firm profilesQuantNet · 25 h · Free
- practice Open-source contribution to a backtesting or quant library as a public track recordGitHub (QuantConnect LEAN, vectorbt, statsmodels) · 50 h · Free
Best certifications for a Quantitative Analyst
Which ones matter, what they cost, and how often people pass.
No certification is required, and none of them will get a Quantitative Analyst hired at a fund. The Financial Risk Manager (FRM) is the one with real screening value, and only for bank market risk, model validation and strat seats: about $1,600 in GARP fees for both parts with early registration, 250 recommended study hours per part, and pass rates of roughly 45-50% on Part I and 50-60% on Part II. The CFA Program is a curriculum in public-markets investment analysis rather than a quant credential - it costs $3,520 for all three levels at early registration and funds do not screen on it - and a master's in financial engineering is a degree, not a certification, at $60,000 to $120,000 in tuition.
Financial Risk Manager (FRM), Parts I and II
GARP (Global Association of Risk Professionals)
- Cost
- $400 one-time enrollment fee plus $600 early / $800 standard per part; about $1,600 total with early registration for both parts
- Study
- About 240 hours per part per GARP's own guidance
- Pass rate
- Approximately 45-50% per part
CFA Program, Levels I-III
CFA Institute
- Cost
- $3,520 (all three levels, early registration) to $4,570 (standard); the one-time $350 enrollment fee was eliminated for exams from February 2026
- Study
- 300+ per level
- Pass rate
- Recent windows: Level I 39-52%, Level II 43-60%, Level III 49-59%; ten-year averages about 40% / 45% / 50%
Master of Financial Engineering / Computational Finance (MFE, MSCF, MAFN)
Universities (Baruch, Carnegie Mellon, Princeton, Berkeley, Columbia, NYU Courant, Chicago)
- Cost
- Roughly $60,000-$120,000 in tuition depending on programme, plus forgone income
- Study
- 1,200-2,000 over 12-24 months
- Pass rate
- Admission rates at top programmes are commonly in the 10-20% range; compare published placement data via the QuantNet rankings
Machine Learning Specialization (Stanford Online / DeepLearning.AI)
Coursera
- Cost
- $49/month subscription, or included in Coursera Plus ($59/month or $399/year); free to audit
- Study
- 95 hours across 3 courses (33 + 34 + 28)
- Pass rate
- Not applicable
Skills employers screen for
Soft skills that decide offers: Scepticism toward your own results - treating a good backtest as a hypothesis, not a finding, Clear technical writing and the ability to defend a method to a sharp critic, Intellectual honesty about negative results, which are most results, Emotional stability when a live strategy loses money for weeks, Speed and composure in timed, high-pressure quantitative interviews, Collaboration with engineers and traders who care about different constraints, Comfort being measured on outcomes rather than effort.
Best courses for a Quantitative Analyst
Checked on the provider's page on 16 September 2026. Some links are affiliate links.
We list 6 courses for Quantitative Analyst work, checked on the provider's page. Use the listed courses to close named gaps rather than as a path: the Mathematics for Machine Learning and Machine Learning specialisations are worth their roughly 165 combined hours if your statistics is thin, and the MIT and Columbia material covers derivatives and risk for free. No course sequence substitutes for the master's degree or the public research track record that actually gets a Quantitative Analyst CV read.
Browse more Quantitative Analyst courses on Coursera Coursera Plus covers most of the courses above for one monthly fee.
Quantitative Analyst interview questions and format
A Quantitative Analyst hiring loop is typically four to six stages over four to eight weeks: a timed online assessment on probability and mental arithmetic, a technical phone screen on statistics and programming, an algorithmic coding round, a research take-home of four to twelve hours, then an onsite covering probability, market-making games and a defence of your own research. The timed online assessment filters the most candidates, because proprietary trading firms cut on it before anyone reads your CV.
Usually four to six stages: an online timed assessment (probability, mental maths, sometimes a coding challenge); a technical phone screen on statistics and programming; a coding round on algorithms and data manipulation; a research take-home (a dataset, a prediction task, a write-up, typically 4-12 hours); and a final onsite with probability, market-making games, a deep discussion of your own research and a fit round. Prop shops weight timed mental-maths and market-making games heavily; hedge funds weight the research take-home; banks weight derivatives pricing and stochastic calculus.
Questions that come up
- You flip a fair coin until you see two heads in a row. What is the expected number of flips?
- Derive the Black-Scholes PDE from a delta-hedged portfolio, and explain each assumption you used.
- You have 1,000 candidate signals and test each at the 5% level. How many will look significant by chance, and what do you do about it?
- Why does standard k-fold cross-validation fail on financial time series, and what would you use instead?
- Your backtest shows a Sharpe of 3.0. List every reason it might be wrong, in order of likelihood.
- Make me a two-sided market on the number of gas stations in Manhattan. Now I lift your offer - what do you do?
- How would you estimate the market impact of a trade that is 5% of average daily volume?
- Explain the difference between implied and realised volatility, and how you would trade a persistent gap between them.
- Write code to compute a rolling z-score over a large tick dataset efficiently. Now make it handle missing data and late-arriving ticks.
- Walk me through a research project of yours. What was your null hypothesis, and how did you try to disprove your own result?
Prep
- A Practical Guide To Quantitative Finance Interviews (Xinfeng Zhou, the 'Green Book')
- Quant Job Interview Questions and Answers (Mark Joshi)
- Advances in Financial Machine Learning (Lopez de Prado) - for backtest validation questions
- QuantNet forums - interview reports by firm
- Levels.fyi quant compensation data for offer negotiation
Quantitative Analyst FAQ
Do I need a PhD to become a Quantitative Analyst at a hedge fund in 2026?
For research roles at systematic hedge funds, a PhD is the most common profile but not a requirement - a strong master's in financial engineering, statistics or CS is a well-trodden alternative, and prop trading firms such as Jane Street, Optiver and IMC hire undergraduates directly on the strength of timed tests. For quant developer, execution and bank strat roles, a bachelor's or master's plus strong engineering is normal. What is non-negotiable is demonstrable mathematical ability under time pressure.
Is a master's in financial engineering worth $60,000 to $120,000 for a career changer?
It depends entirely on the programme's published placement data. Top programmes (Baruch, Carnegie Mellon MSCF, Princeton, Berkeley, Columbia, NYU Courant) publish employment-at-three-months and salary figures and run their own recruiting relationships, and for a career changer that pipeline is most of the value. A programme without published placement data is a $90,000 bet on a brochure. Compare via the QuantNet rankings before you apply, and check whether the alternative - lateralling into a quant developer or data role and converting internally - is open to you, since it costs nothing.
How much does a Quantitative Analyst actually make in year one at a fund versus a bank?
It splits sharply by firm. First-year total compensation at top funds and prop shops is commonly reported at $250k-$450k, with Citadel graduate quant researchers in Chicago reported at $275k-$475k and Jane Street paying interns the annualised equivalent of about $300k. At banks it is far lower - levels.fyi puts the Goldman Sachs quantitative researcher median near $150k and JPMorgan near $217k. The all-firm median on levels.fyi is around $200k-$220k, which is the honest number for the broad market rather than the headline tier.
Will the CFA or the FRM help me get a Quantitative Analyst job in 2026?
The CFA will not - it covers public-markets investment analysis, not stochastic calculus or statistical learning, and funds do not screen on it. The FRM is genuinely useful for bank market-risk, model-validation and strat roles, costs about $1,600 in GARP fees with early registration ($400 enrollment plus $600 per part), takes roughly 240 hours per part, and has pass rates around 45-50%. For buy-side research, a public research track record and a strong coding portfolio outrank both.
I am a software engineer at 35: what is the fastest route into a Quantitative Analyst role?
Lateral, not retrain. Apply for quantitative developer, execution engineering, low-latency systems or research-platform roles at funds and banks - these hire on engineering strength with a lower mathematical bar, pay well immediately, and put you inside the building. Meanwhile close the statistics gap (Andrew Ng's Machine Learning Specialization, 95 hours; Lopez de Prado on backtest validation) and contribute to signal research internally. Firms convert developers to researchers regularly, usually over 18-36 months, and it costs you nothing in tuition.
Can I get hired as a Quantitative Analyst on a self-taught trading track record alone?
Rarely, and only if the research is genuinely rigorous. A profitable backtest is not evidence; firms will ask about purged cross-validation, multiple-testing correction, transaction costs, capacity and out-of-sample discipline, and a track record that crumbles under those questions is worse than none. Where self-study does pay off is as a supplement: a QuantConnect paper-trading record over 6-12 months, open-source contributions to a backtesting library, or a well-written research note gives you something concrete to discuss in a research interview.
What does a Quantitative Analyst actually do all day at a hedge fund?
Mostly data work and failed experiments. A typical day starts by checking overnight performance of the strategies you own, then a short research meeting with blunt methodology questions, then three hours cleaning a dataset, aligning timestamps to avoid look-ahead and handling survivorship in the security master. The afternoon goes on fitting and validating a candidate signal with purged, embargoed cross-validation, applying a multiple-testing correction and watching most of the apparent edge disappear, then writing it up for a senior researcher who will try to break it. Roughly 90 to 95 percent of research ideas fail out of sample.
Is Quantitative Analyst a realistic career change for someone switching at 40?
Rarely, and only from a technical starting point. The mathematics bar is real and this page's own estimate is 24 to 48 months of preparation from an unrelated background, usually including a master's degree that costs $60,000 to $120,000 in tuition plus forgone income. At 40 the honest version is the lateral route: if you already write production code, move into a quantitative developer, execution engineering or data-platform seat at a fund or bank, get paid while you close the statistics gap, and convert to research internally over 18 to 36 months. Starting from a non-technical job at 40 is a very long shot.
Quantitative Analyst versus investment banking analyst: what is the difference in pay and hours?
Different jobs that share an industry. A Quantitative Analyst is measured on whether a model makes money and works roughly 45 to 60 hours a week; an investment banking analyst is measured on client deliverables and commonly works 70 to 90. Pay at the top of each is comparable, with first-year total compensation at leading funds and proprietary shops commonly reported at $250,000 to $450,000, but the routes diverge completely: banking recruits on target-school internships and modelling tests, while quant hiring runs on timed probability tests and a research take-home, which is why it is closer to a meritocracy of demonstrable ability.
How much does a Quantitative Analyst earn in New York City versus the national median?
About 28 percent more on paper, but firm type matters far more than city. Applying the Salary Roadmap New York City multiplier of 1.28 to the Levels.fyi all-firm quantitative researcher median of roughly $210,000 gives about $269,000. Compare that with the spread by employer in the same dataset: Citadel around $600,000, Two Sigma around $420,000, JPMorgan around $217,500 and Goldman Sachs around $150,750. Chicago (1.05) hosts Citadel and Jump, and New York hosts most of the rest, so the metro premium is largely a composition effect - you are paid for the seat, not the postcode.
How long does it take to become a Quantitative Analyst while working full time?
Plan on 24 to 48 months. The roadmap on this page totals about 1,380 study hours, which is roughly 18 months at 15 to 20 hours a week, and that only gets you to the point where interviews are winnable - it does not include the master's degree that most career changers still need, or the 18 to 36 months an internal conversion from a quantitative developer seat takes. The cheapest honest test is step one: if you cannot compute a conditional expectation for a joint density after 250 hours of work, the switch is probably not realistic.
Sources
Every number on this page traces to one of these. Page checked 16 September 2026.
- levels.fyi/t/data-scientist/title/quantitative-researcher
- levels.fyi/companies/citadel/salaries/data-scientist/title/quantitative-researcher
- levels.fyi/companies/two-sigma/salaries/data-scientist/title/quantitative-researcher
- levels.fyi/companies/goldman-sachs/salaries/data-scientist/title/quantitative-researcher
- quantt.co.uk/resources/quant-researcher-salary-guide
- bls.gov/ooh/math/mathematicians-and-statisticians.htm
- bls.gov/ooh/business-and-financial/financial-analysts.htm
- garp.org/frm/fees-payments
- garp.org/frm
- cfainstitute.org/programs/cfa-program/dates-fees
- coursera.org/specializations/financialengineering
- coursera.org/specializations/machine-learning-introduction
- coursera.org/specializations/machine-learning-trading
- quantconnect.com/pricing
- lucylabs.gatech.edu/ml4t/
- ocw.mit.edu/courses/18-s096-topics-in-mathematics-with-applications-in-finance-fall-2013/
- quantnet.com/mfe-programs-rankings/
- quantnet.com/forum/