Quant Trading and Research terms
- 48
- terms
Quant25
- Alpha
- Return beyond what market exposure or a benchmark explains. A fund that earns 12% when its beta-adjusted benchmark earns 9% has about 3% of alpha.ExampleA fund with a beta of 1.2 earns 14% when the market returns 10% and the risk-free rate is 4%. Its expected return is 4% + 1.2 x 6% = 11.2%, so its alpha is 2.8%.See also Beta, Sharpe ratio, Factor
- Arbitrage
- Locking in a riskless profit from prices that are inconsistent with each other, by buying the cheap side and selling the rich one at the same time. True arbitrage is rare and quickly competed away; most trades called arbitrage carry some risk.ExampleA share trades at $40.00 on one exchange and $40.05 on another. Buying on the first and selling on the second locks in $0.05 a share before fees.See also Statistical arbitrage, Put-call parity, Market maker
- Backtest
- Running a trading strategy on historical data to see how it would have done. Useful, but easy to fool yourself with: trading costs, slippage, and data you would not have had at the time all have to be handled honestly.ExampleA strategy backtests at 12% a year before costs but turns over its whole book every day. At 3 bps of cost per turnover, 250 trading days cost 7.5% a year, and most of the edge is gone.See also Overfitting, Sharpe ratio
- Bayes' theorem (Bayes)
- The rule for updating a probability with new evidence: P(A given B) = P(B given A) x P(A) / P(B). It explains why a positive result on a rare condition can still be likely to be false.ExampleA condition affects 1% of people; a test catches 99% of cases and wrongly flags 1% of healthy people. Given a positive test, the chance of having it is 0.0099 / (0.0099 + 0.0099) = 50%.See also Conditional probability, Expected value
- Black-Scholes model (Black-Scholes-Merton)
- The classic formula for pricing European options from the stock price, strike, time to expiry, risk-free rate, and volatility. Its assumptions, such as constant volatility and no jumps, fail in practice, but it is the common language of options markets.ExampleA one-year at-the-money call on a $100 stock with 20% vol and a 0% rate is worth about $8. A quick check: 0.4 x price x vol x the square root of time = 0.4 x 100 x 0.2 x 1 = $8.See also Implied volatility, Call option, Monte Carlo simulation
- Conditional probability
- The chance of an event once you know another has happened: P(A given B) = P(A and B) / P(B). Most quant brainteasers turn on spotting what you have been told and shrinking the sample space to it.ExampleTwo dice: given the first die is even, what is the chance the sum is 8? Of 18 equally likely pairs, three work (2 and 6, 4 and 4, 6 and 2), so 3/18 = 1/6.See also Bayes' theorem, Expected value
- Correlation
- A number from -1 to 1 for how closely two quantities move together. Below 1, combining assets lowers portfolio risk, which is the arithmetic behind diversification.ExampleTwo assets each with 20% volatility, held 50/50 with zero correlation, give a portfolio volatility of the square root of (0.25 x 0.04 + 0.25 x 0.04), about 14.1%, not 20%.See also Standard deviation, Beta, Linear regression
- Expected value
- The probability-weighted average outcome. A bet that pays $10 with probability 0.3 and nothing otherwise has an expected value of $3. Quant interviews test it constantly with dice, cards, and betting games.ExampleRoll one die and get paid the number in dollars: the expected value is (1 + 2 + 3 + 4 + 5 + 6) / 6 = $3.50. Pay $3 to play and you have a 50 cent edge per roll.See also Kelly criterion, Market-making game
- Factor
- A characteristic that helps explain returns across many securities, such as value, momentum, size, or quality. Factor models split a portfolio's return into factor exposures plus what is left over, which is the alpha.ExampleA portfolio returns 9%. A factor model attributes 6% to the market, 1.5% to a tilt toward value stocks, and 1% to momentum, which leaves 0.5% of alpha.See also Alpha, Statistical arbitrage
- High-frequency trading (HFT)
- Automated trading that holds positions for milliseconds to seconds and earns small edges on enormous volume. Speed, hardware, and servers placed next to the exchange's matter as much as the model.ExampleA firm earns a tenth of a cent a share on 50m shares a day. That is $50,000 a day, about $12.5m a year over 250 trading days, as long as it stays fast enough to win the trades.See also Market maker, Bid-ask spread
- Kelly criterion
- A rule for how much of your bankroll to bet when you have an edge. For an even-money bet you win with probability p, bet 2p - 1 of your bankroll, so 10% when p = 0.55. Many traders bet a fraction of Kelly to cut volatility.ExampleYou win an even-money bet 60% of the time. Kelly says bet 2 x 0.6 - 1 = 20% of the bankroll, so $20,000 of $100,000; half Kelly would be $10,000.See also Expected value
- Linear regression (regression)
- Fitting a straight line y = a + b x to data by minimizing the sum of squared errors. The slope b says how much y moves per unit of x on average. It is how beta is estimated and the first model most quant researchers try.ExampleRegressing a stock's daily returns on the market's gives a slope of 1.3: on average the stock moves 1.3% for each 1% move in the market.See also Beta, Correlation, Overfitting
- Market-making game
- A quant trading interview exercise: the interviewer asks you for a two-sided market on an uncertain number, like the sum of three dice, and trades against you. It tests estimation, updating on new information, and risk control under pressure.ExampleAsked for a market on the sum of three dice, you quote 9 at 12 around the 10.5 expected value. Told one die shows a 6, the expected sum is 6 + 7 = 13, so you move to 12 at 14.See also Market maker, Expected value, Quant trader
- Mean reversion
- The tendency of a price, spread, or ratio to drift back toward its average after moving away. Pairs trading and many statistical arbitrage strategies are bets on it.ExampleTwo similar bank stocks usually trade at a price ratio of 1.5. When it stretches to 1.7, a pairs trader shorts the expensive one and buys the cheap one, betting the ratio falls back toward 1.5.See also Statistical arbitrage, Factor
- Monte Carlo simulation
- Estimating something by running thousands of random scenarios and averaging the results. It is used to price complex derivatives and to see how a portfolio behaves across many possible paths.ExampleTo price an option that pays on a stock's average price over a year, simulate 100,000 price paths, compute the payoff on each, average them, and discount. More paths, less noise.See also Black-Scholes model, Expected value
- Order book (limit order book)
- The live list of resting buy orders (bids) and sell orders (offers) for a security, by price and size. A market order fills against the best prices first and walks down the book if it is larger than the size shown there.ExampleBest offer 100.02 for 300 shares, next 100.05 for 1,000. A market buy for 400 fills 300 at 100.02 and 100 at 100.05.See also Bid-ask spread, Market maker, Slippage
- Overfitting (curve fitting)
- Fitting a model so closely to past data that it learns the noise instead of the signal, so it fails on new data. Out-of-sample testing and fewer parameters are the usual defenses.ExampleA model with 40 parameters shows a Sharpe of 3.0 on ten years of data, then 0.2 on the next year it never saw. A 4-parameter version shows 1.1 in sample and 0.9 out of sample, and it is the one to trade.See also Backtest, Quant researcher
- Put-call parity
- For European options with the same strike and expiry: call minus put equals the stock price minus the present value of the strike. If prices break it, a riskless trade locks in the difference.ExampleStock at 100, strike 100, rates at zero, so call and put should cost the same. If the call is 7 and the put 5, sell the call, buy the put, and buy the stock for 98: the package is worth 100 at expiry whatever happens.See also Call option, Put option, Arbitrage
- Quant researcher
- The person who finds and tests trading signals using statistics, machine learning, and large datasets. Many roles prefer graduate degrees in math, statistics, physics, or CS, though several firms also hire undergraduates.ExampleA researcher tests whether stocks that miss earnings keep drifting lower for 60 days. She builds the signal on 2005 to 2018 data, tests it on 2019 onward, and hands it to trading only if it holds.See also Quant trader, Backtest, Alpha
- Quant trader
- A trader who runs systematic or semi-systematic strategies: setting parameters, managing risk, and making quick calls with models in hand. Firms like Jane Street, SIG, and Optiver hire undergraduates straight into this seat.ExampleA trader runs an options market-making strategy. When a large order moves implied vol 3 points in a minute, she decides whether to widen quotes, cut size, or hedge with the underlying.See also Quant researcher, Market-making game, Market maker
- Random walk
- A path built from independent random steps. Its expected position stays where it started, but its spread grows with the square root of the number of steps. It is the simplest model of a price with no predictable direction.ExampleFlip a fair coin 100 times, gaining $1 on heads and losing $1 on tails. The expected result is $0 and the standard deviation is the square root of 100, or $10.See also Standard deviation, Mean reversion, Monte Carlo simulation
- Slippage
- The gap between the price a trade was expected to get and the price it actually got, from the spread, market impact, or prices moving before the fill. Backtests that ignore it overstate returns.ExampleA strategy expects to buy 10,000 shares at $50.00 and averages $50.06. Slippage is $0.06 a share, or $600 on the trade.See also Order book, Backtest, Bid-ask spread
- Standard deviation (variance)
- How widely outcomes spread around their average: the square root of the variance, which is the average squared distance from the mean. For returns it is what traders call volatility, and it scales with the square root of time.ExampleA stock with a 1% daily standard deviation has about 1% x the square root of 252, or roughly 16%, annual volatility.See also Correlation, Sharpe ratio, Implied volatility
- Statistical arbitrage (stat arb)
- Trading many small, statistically predicted mispricings across hundreds of securities at once, usually with little net market exposure. Each trade is close to a coin flip; the edge only shows up across thousands of them.ExampleA book holds 500 longs and 500 shorts, each a small bet that is right 51% of the time. Any single trade is noise; across 100,000 trades a year the edge is steady and market moves mostly cancel.See also Mean reversion, Factor, High-frequency trading
Trading and products10
- Bid-ask spread
- The gap between the highest price a buyer will pay (the bid) and the lowest a seller will accept (the ask). A stock quoted 100.00 bid, 100.02 ask has a 2 cent spread, which is roughly what a market maker earns for providing liquidity.ExampleA bond quoted 99.50 bid, 99.75 ask: buying $10m face costs $9,975,000 and selling it straight back brings $9,950,000. Crossing the spread both ways cost $25,000.See also Market maker, High-frequency trading
- Call option
- The right, but not the obligation, to buy an asset at a set strike price. A call with a $50 strike on a stock trading at $60 at expiry is worth $10, minus whatever premium you paid for it.ExamplePay $3 for a call with a $50 strike. With the stock at $58 at expiry, it is worth $8 and you net $5; below $50 it expires worthless and you lose the $3. Breakeven is $53.See also Put option, Delta, Black-Scholes model
- Convexity
- How much a bond's duration changes as yields move, the curve in the price and yield relationship. A plain bond has positive convexity: it gains more when yields fall than it loses when they rise by the same amount.ExampleA bond with a duration of 10 and positive convexity might gain about 10.6% when yields fall 1% and lose only about 9.4% when they rise 1%. Duration alone predicts 10% both ways.See also Duration, DV01
- Delta
- How much an option's price changes for a $1 move in the underlying. A call with a delta of 0.5 gains about 50 cents when the stock rises $1. Traders hedge by holding the opposite delta in the underlying.ExampleYou sell 100 call contracts, 10,000 shares' worth, with a delta of 0.40. To be flat you buy 4,000 shares; if delta rises to 0.55, you buy 1,500 more.See also Gamma, Hedging, Call option
- Futures contract (futures)
- A standardized, exchange-traded agreement to buy or sell an asset at a set price on a future date. Gains and losses settle daily through margin, which makes futures cheap to trade and convenient for hedging.ExampleBuy one crude oil contract, 1,000 barrels, at $70. If it settles at $72 that day, $2,000 goes into your margin account; if it settles at $68, $2,000 comes out.See also Hedging, Repo
- Gamma
- How much an option's delta changes for a $1 move in the underlying. Gamma is highest for options near the strike close to expiry, which means their hedges need frequent adjusting.ExampleA call has a delta of 0.50 and a gamma of 0.05. If the stock rises $2, delta goes to about 0.60, so a trader long the call and short stock at 0.50 must sell more stock to stay hedged.See also Delta, Implied volatility
- Hedging
- Taking an offsetting position to reduce a risk you do not want. A trader who owns a corporate bond can sell Treasury futures so that a rate move hits both sides and roughly cancels, leaving mostly the credit risk.ExampleAn airline will buy 10m gallons of jet fuel in six months and fears a price rise. It buys fuel futures today; if prices climb $0.50 a gallon, the futures gain about $5m to offset the higher bill.See also Delta, DV01, Futures contract
- Implied volatility (implied vol)
- The volatility the market is pricing into an option, backed out of the option's price with a model like Black-Scholes. Options traders think of themselves as buying and selling volatility, so they quote in vol rather than dollars.ExampleA call priced at $4.00 implies 30% vol. The next day, stock unchanged, it trades at $4.60, implying 35%. Traders say vol went up 5 points, not that the option rose 60 cents.See also Black-Scholes model, VIX, Gamma
- Market maker
- A firm that continuously quotes prices to buy and sell, earning the bid-ask spread while managing whatever inventory it ends up holding. Electronic firms like Citadel Securities and Jane Street and bank flow desks all make markets.ExampleA market maker quotes a stock at 50.00 bid, 50.02 ask. Buying 10,000 shares from one client and selling them to another earns $200; if the price drops before it sells, the inventory loss can wipe that out.See also Bid-ask spread, Market-making game, High-frequency trading
- Put option
- The right, but not the obligation, to sell an asset at a set strike price. A put with a $50 strike on a stock trading at $40 at expiry is worth $10. Investors buy puts as insurance against a drop.ExampleOwn 1,000 shares at $80 and buy $75 puts for $2 a share. If the stock falls to $60, the puts pay $15 a share, capping your loss at $7 a share including the premium.See also Call option, Hedging
Recruiting10
- Coffee chat (informational interview)
- A short call or meeting, usually 15 to 30 minutes, with someone at a firm to learn about their job. It is the core of networking, and a good one ends with a referral or another name to contact.ExampleA sophomore emails a second-year analyst from her school, gets 20 minutes on Thursday, asks three questions about the group, sends a thank-you that night, and is introduced to a VP the next week.See also Networking, Fit interview
- Diversity program
- An early insight program or internship track for students from underrepresented groups. Many open earlier than the main summer analyst program, some to freshmen and sophomores, and some lead straight to an internship offer.ExampleA bank's sophomore diversity program takes applications in the fall, runs a two-day event in spring, and gives strong participants an early interview for the junior summer analyst class.See also Sophomore program, Summer analyst
- Exploding offer
- An offer with a short deadline, sometimes only days, meant to get you to decide before you hear from other firms. Asking politely for more time is normal; the worst answer is usually just no.ExampleA bank makes an offer on Monday with a Friday deadline, and another bank's superday is next Tuesday. Telling the second bank about the deadline often gets its process moved up.See also Return offer, On-cycle
- Fit interview (behavioral interview)
- The questions about you: walk me through your resume, why this firm, why this path, a time you led or failed. Every finance process has them, and weak answers sink otherwise strong technical candidates.ExampleWhy investment banking? A strong answer names something you did (a valuation for your investment club), what you liked about the work, and why this seat gives you more of it.See also Technical interview, Superday, HireVue
- On-campus recruiting (OCR)
- The process where firms post jobs and interview through your school's career center. Schools where firms recruit heavily get the most slots; elsewhere, networking and direct applications carry more weight.ExampleAt a school with OCR, a bank posts its role on the career portal, collects resumes, and runs first rounds in campus interview rooms. At a school without it, you apply online and lean on referrals.See also Networking, Summer analyst
- Online assessment (OA)
- A timed online test sent after you apply: numerical reasoning, logic games, or personality questions for banks, and math or coding problems for quant roles.ExampleA bank's online assessment gives you 20 minutes for 18 numerical questions, such as working out a company's year-on-year revenue growth from a bar chart.See also HireVue, Superday
- Return offer
- A full-time job offer given to an intern at the end of the summer. At most banks it is the main path into a full-time analyst job, so the internship works like a ten-week interview.ExampleA class of 80 summer analysts is reviewed in August. 65 return offers is an 81% rate, and the 15 without one recruit again in the fall.See also Summer analyst, Exploding offer
- Rolling admissions (rolling applications, rolling basis)
- Reviewing applications as they arrive and filling seats before the posted deadline. Applying in the first week or two after a program opens can matter more than the deadline itself.ExampleA program opens January 10 with a March 1 deadline. If most interview slots fill by mid-February, an application sent February 20 arrives after most of the decisions are made.See also Summer analyst, Online assessment
- Sophomore program
- An early program for second-year students, such as an insight week, a spring program, or a sophomore internship. Some feed directly into the junior summer analyst process, which now recruits a year or more ahead.ExampleA bank runs a three-day spring insight program for 60 sophomores. Strong participants get fast-tracked into interviews for the junior summer analyst class.See also Diversity program, Summer analyst
- Technical interview
- Questions that test finance knowledge: accounting, valuation, DCF, and LBO for banking and PE; markets for trading; probability and mental math for quant. Expect prompts like walk me through a DCF, or how does $10 of depreciation flow through the three statements.ExampleDepreciation rises $10 at a 25% tax rate. Net income falls $7.50; cash rises $2.50 after adding back the $10; PP&E falls $10; assets fall $7.50 and so does retained earnings.See also Fit interview, Discounted cash flow, Paper LBO
Markets and macro2
- Basis point (bps)
- One hundredth of a percentage point, or 0.01%. A rate that moves from 4.25% to 4.50% has risen 25 basis points.ExampleA loan priced at SOFR + 350 bps with SOFR at 4.30% pays 7.80%. On a $100m loan, each basis point is worth $10,000 a year.See also DV01, Credit spread
- VIX
- Cboe's index of expected 30-day volatility in the S&P 500, calculated from option prices. It tends to jump when stocks fall, which is why people call it the fear gauge.ExampleA VIX of 16 means options price the S&P 500 to move about 16% a year, or roughly 1% a day, since 16 divided by the square root of 252 trading days is about 1.See also Implied volatility, Risk-on and risk-off
Valuation1
- Beta
- How much a stock tends to move relative to the market. A beta of 1.2 means the stock has historically moved about 1.2% for each 1% move in the market. It feeds the cost of equity through CAPM.ExampleA stock with a beta of 0.8 would be expected to rise about 4% when the market rises 5%. Utilities often sit below 1; fast-growing tech names often sit above it.See also Capital asset pricing model, Cost of equity, Alpha