Reference

The quant vocabulary, in plain English.

Every quant and trading term you'll bump into, explained without equations. Where a term has a deeper write-up, follow the Learn more link into the concept library.

A B C D E F G H IJK L M N O P Q R S T U V WXY Z

313 terms

A

Adverse selectionMicrostructure: The market maker's nightmare: the traders most eager to hit your quote are often the ones who know something you don't, so you tend to lose exactly when it matters. To survive, market makers widen their spreads to cover it. Learn more →
AlphaPerformance: The slice of return that comes from genuine skill or edge, beyond what the market handed you for free. Positive alpha is the holy grail every active manager chases. Learn more →
Alpha decayQuant research: The slow fading of a strategy's edge as more people discover it, markets change, or the pattern simply stops working. Almost every profitable idea decays eventually, which is why quants keep researching new ones. Learn more →
American optionOptions: An option you can exercise any time up to expiry, not just on the final day. More flexible, so usually a touch more expensive.
Annualized ReturnPerformance: A return stretched or squeezed to express it as a per-year figure, so periods of different lengths can be compared fairly. It is the standard way to state performance.
ArbitrageTrading: Making low-risk profit by exploiting a price difference for the same thing in two places, buy cheap here, sell dear there. These gaps are usually tiny and vanish fast.
Ask (Offer)Trading: The lowest price a seller is currently willing to accept. If you want to buy instantly, this is the price you would pay.
Asset AllocationPortfolio: Deciding what slice of your money goes into each broad type of investment, like stocks versus bonds versus cash. It is usually the single biggest driver of how a portfolio behaves over time.
Asset ClassMarkets: A broad family of investments that behave similarly, such as stocks, bonds, or commodities. Mixing different asset classes helps balance risk.
AssignmentOptions: When an option buyer exercises and the seller (the writer) is picked to fulfil the contract. If you wrote a call, you now must sell the stock at the strike whether you like it or not.
At the moneyOptions: An option whose strike is roughly equal to the current market price. It sits right on the fence between worthwhile and worthless.
AutocorrelationTime Series: When a series is related to its own past values, so today's number carries a clue about tomorrow's. Strong autocorrelation can hint at a tradable pattern, or it can just mean your data points are not as independent as you assumed. Learn more →

B

BacktestQuant research: A dress rehearsal that runs your strategy on past market data to see how it would have performed. Useful for spotting bad ideas, but a great backtest is no promise of future profits. Learn more →
BackwardationDerivatives: When futures for later delivery cost less than the price today, so the price curve slopes downward. It often signals tight, in-demand supply right now.
Barrier optionOptions: An exotic option that only springs to life, or dies, if the price touches a set level. It is cheaper than a normal option because it might never activate at all. Learn more →
BasisDerivatives: The gap between an asset's current cash price and its futures price. Traders watch the basis shrink toward zero as the futures nears expiry.
Bayes' theoremProbability: A rule for flipping a probability around: it turns the chance of the evidence given a cause into the chance of the cause given the evidence. In plain terms, it tells you how much to revise your beliefs after seeing new data. Learn more →
Bayesian vs frequentistProbability: Two schools of statistics. Frequentists treat probability as long-run frequency and the unknown truth as fixed; Bayesians treat probability as a degree of belief that gets updated with data. They often reach similar answers by different routes. Learn more →
Bear MarketMarkets: A stretch of time when prices are generally falling and people feel gloomy, often defined as a drop of 20% or more. The nickname comes from a bear swiping its paws downward.
BenchmarkMarkets: A yardstick you compare your performance against, often an index like the S&P 500. If your investments beat the benchmark, you did better than the average.
BetaPerformance: How strongly an investment moves with the overall market. A beta of one moves in step, above one amplifies the market's swings, and below one is more muted. Learn more →
BiasStatistics: A systematic tendency for an estimate or model to miss the truth in the same direction over and over. Unlike random noise, more data alone will not wash bias away. Learn more →
Bias-variance tradeoffRegression: The central tension in modeling: too simple a model is biased and misses the pattern, while too flexible a model swings wildly with the noise. Good models find the sweet spot between the two. Learn more →
BidTrading: The highest price a buyer is currently willing to pay for something. If you want to sell instantly, this is the price you would get.
Bid-ask spreadMicrostructure: The gap between the highest price buyers will pay and the lowest price sellers will accept. It is the toll you pay to trade instantly, and it is the market maker's basic source of profit.
Binomial distributionDistributions: The distribution of how many successes you get in a fixed number of yes-or-no trials, like the number of heads in ten coin flips. It needs a fixed count of trials and a steady success chance each time. Learn more →
Black-Litterman ModelPortfolio: A smarter way to set portfolio weights that blends the market's default view with your own opinions. It avoids the wild, extreme bets that plain optimization tends to produce. Learn more →
Black-Scholes modelOptions: The famous formula that estimates a fair price for an option from the current price, strike, time left, interest rate, and volatility. It gave traders a shared language for pricing options. Learn more →
Blue ChipInstruments: A large, well-known, financially solid company with a long track record, like Coca-Cola or Microsoft. Blue chips are seen as safer, steadier bets.
BondInstruments: A loan you make to a company or government. They pay you interest for a set period and then give your money back at the end.
BootstrapStatistics: A trick for gauging how reliable an estimate is by resampling your own data, with replacement, over and over and watching how much the answer varies. It fakes many studies out of the one dataset you have. Learn more →
BrokerParticipants: A company that lets you buy and sell investments, acting as your gateway to the market. You place orders with them and they get them executed.
Bull MarketMarkets: A stretch of time when prices are generally rising and people feel optimistic. The nickname comes from a bull attacking by thrusting its horns upward.
Buy-SideParticipants: The part of the industry that invests money, like hedge funds, pension funds, and mutual funds. They buy and hold investments to grow their clients' money.

C

CAGRPerformance: The single steady yearly growth rate that would take you from your starting value to your ending value. It smooths the bumpy real path into one clean average.
Call optionOptions: The right (not the obligation) to BUY something later at a price fixed today. Like a coupon that lets you buy a stock at $100 anytime this month, valuable if the stock climbs above $100.
Call spreadStrategies: Buying one call and selling another at a higher strike. This caps both your cost and your maximum gain, a cheaper, limited bet that the price rises.
Calmar RatioPerformance: A performance score that stacks a strategy's return against its worst drawdown. It rewards returns that come without gut-wrenching crashes along the way. Learn more →
CapacityQuant research: The most money a strategy can manage before its own trading moves prices enough to eat the profits. A brilliant idea that only works with small size has low capacity and cannot run a big fund. Learn more →
Capital at RiskRisk: The portion of your money genuinely exposed to loss on a given trade or strategy, as opposed to what is parked safely. Keeping this modest is the core of staying in the game.
Capital GainTrading: The profit you make when you sell something for more than you paid. Buy a stock at $10 and sell at $15, and your capital gain is $5.
CAPMPerformance: A classic model that says an asset's fair expected return depends only on how much market risk it carries. It is the textbook starting point for pricing risk, even if reality is messier. Learn more →
Central limit theoremProbability: When you add up or average lots of small independent random effects, the result tends to look like a bell curve, even if the individual pieces do not. It is why the normal distribution shows up almost everywhere. Learn more →
ClearingMarkets: The behind-the-scenes process that confirms and organizes a trade so it can settle smoothly. A clearinghouse sits in the middle to make sure both sides deliver.
Coefficient (slope)Regression: The number in a regression that says how much the outcome changes when an input rises by one unit. A bigger coefficient means that input has a stronger pull on the result. Learn more →
Coherent Risk MeasuresRisk: A checklist of common-sense rules a good risk measure should obey, such as rewarding diversification. Value at Risk famously breaks one of them, which is why Expected Shortfall is often preferred. Learn more →
CointegrationQuant research: A tight statistical bond where two prices can wander individually but tend to stay tethered over time, like a dog on a leash. It is the property that makes pairs trading work, because the gap between them keeps snapping back. Learn more →
ColocationMicrostructure: Renting space for your servers right next to the exchange's own computers to shave precious microseconds off your reaction time. In the speed race of high-frequency trading, physical distance to the exchange is money. Learn more →
CommoditiesInstruments: Raw physical goods that are traded, like oil, gold, wheat, or coffee. Their prices swing with supply and demand around the world.
CompoundingPerformance: Earning returns on top of your past returns, so growth snowballs over time. Given enough years it does far more heavy lifting than most people expect.
ConcentrationPortfolio: How much of a portfolio is packed into just a few holdings. Heavy concentration can supercharge returns but leaves you badly exposed if one bet goes wrong.
Conditional probabilityProbability: The chance of one thing happening given that you already know something else happened. It is how you update your odds as new information arrives, like the chance of rain given that the sky is already grey.
Confidence intervalStatistics: A range around an estimate that is meant to capture the true value with a stated level of confidence, such as 95%. It is a way of admitting your estimate is fuzzy and showing how fuzzy. Learn more →
ContangoDerivatives: When futures for later delivery cost more than the price today, so the price curve slopes upward. It often reflects the cost of storing and carrying the asset until then.
ConvexityOptions: The curved, non-straight-line link between an option's value and the underlying price. This bend is why an option can gain more on the way up than it loses on the way down for the same size move.
CorrelationMarkets: How closely two things move together. If two stocks rise and fall in sync, they are highly correlated; if one zigs while the other zags, they move opposite.
Correlation (Portfolio Context)Portfolio: A measure of whether two investments tend to move together, in opposite directions, or independently. Combining things that do not move in lockstep is what makes diversification actually work.
Correlation ClusteringRisk: Grouping assets by how tightly they move together, revealing hidden families within a portfolio. It helps you see that seemingly separate bets are really the same bet in disguise. Learn more →
CounterpartyParticipants: The other party on the opposite side of your trade. Counterparty risk is the danger that they fail to pay or deliver what they promised.
CovarianceStatistics: A measure of whether two variables tend to move in the same direction, positive when they rise together and negative when one rises as the other falls. Correlation is just covariance rescaled to a tidy -1 to +1 range. Learn more →
Covariance EstimationPortfolio: The job of figuring out, from past data, how investments move together, before you can build a portfolio. Getting this wrong is one of the biggest hidden dangers in portfolio construction. Learn more →
Covariance MatrixPortfolio: A big table that captures how every pair of investments in a portfolio moves together. It is the raw material most portfolio math needs to estimate overall risk.
Covered callStrategies: Owning a stock and selling a call option on it to earn extra income from the premium. You give up big upside in exchange for cash now, a bit like renting out a house you already own.
Cross-validationRegression: A way to check a model honestly by training it on part of the data and testing it on the part it has never seen, then rotating. It is your best defense against fooling yourself with overfitting. Learn more →
CryptocurrencyInstruments: Digital money that runs on computer networks instead of banks, like Bitcoin. Prices can move dramatically and it is not backed by any government.
Cumulative distribution functionDistributions: A running total that tells you the chance of a value being at or below any given point. It climbs from 0 up to 1 as you sweep from the smallest possible value to the largest.

D

Data snoopingQuant research: Torturing the data by trying hundreds of ideas until one looks great purely by luck, also called p-hacking. Test enough random patterns and some will appear profitable by chance, then fail the moment you trade them. Learn more →
Decision treeMachine learning: A model that makes predictions by asking a chain of yes/no questions, like a flowchart, until it reaches an answer. Easy to understand, but a single tree is often too crude to be reliable on its own.
Deflated Sharpe ratioQuant research: A more honest performance score that shrinks a strategy's apparent quality to account for how many variations you tried before landing on it. It answers the question: is this edge real, or did I just get lucky after testing a hundred ideas? Learn more →
Degrees of freedomStatistics: Roughly, how many independent pieces of information you have left to estimate things once some have been used up. It shows up as a knob that adjusts many statistical formulas for small samples.
DeltaOptions: How much an option's price moves when the underlying moves by $1. A delta of 0.5 means the option gains about 50 cents for every $1 the stock rises. Learn more →
Delta hedgingOptions: Cancelling out an option's price risk by holding an offsetting amount of the underlying, so small moves wash out. Traders keep adjusting the hedge as the delta drifts. Learn more →
DerivativeInstruments: An investment whose value comes from something else, like a bet based on the price of a stock, oil, or interest rates. Options and futures are common examples.
DiversificationPortfolio: Spreading your money across many different investments so no single bad bet can sink you. It is the old idea of not putting all your eggs in one basket, made a little more scientific.
DividendInstruments: A share of a company's profits paid out to shareholders, usually as cash. It is a way to earn money from a stock even if you never sell it.
DrawdownTrading: How far your account has fallen from its highest point. A 20% drawdown means you are down a fifth from your peak, and it shows how painful a losing streak has been. Learn more →
Drawdown RecoveryRisk: The time and gain it takes to climb back to a previous peak after a loss. Deep holes are punishing, because it takes an outsized gain just to break even.

E

Efficient FrontierPortfolio: The set of portfolios that squeeze out the most expected return for each level of risk. Any portfolio below this curve is wasteful, because you could earn more without taking on more risk. Learn more →
Equal WeightPortfolio: Putting the same amount of money into every holding, regardless of company size. It gives smaller names a real voice and often tilts you toward them. Learn more →
EquitiesInstruments: Just another word for stocks, ownership stakes in companies. When people say they invest in equities, they mean they buy shares.
EstimatorStatistics: A recipe for guessing an unknown quantity from data, like using the sample average to estimate the true average. Good estimators are right on average and do not swing around too much. Learn more →
ETF (Exchange-Traded Fund)Instruments: A basket of many investments bundled into one thing you can buy like a single stock. Buying one share of an ETF spreads your money across everything inside it.
European optionOptions: An option you can only exercise on the expiry date itself, not before. Less flexible than the American style.
Excess ReturnPerformance: The return earned above some baseline, usually a safe cash rate or a benchmark. It is what actually rewards you for taking on risk.
ExchangeMarkets: The marketplace where buyers and sellers meet to trade stocks and other assets, like the New York Stock Exchange or Nasdaq. It matches orders and keeps trading orderly.
ExecutionExecution: The nuts-and-bolts job of actually getting your intended trades filled in the market at good prices. Even a great strategy can bleed money if orders are sent clumsily and push prices against you. Learn more →
ExerciseOptions: Actually using your option, buying (call) or selling (put) at the strike price. You only bother when it is in your favour.
Exotic optionOptions: Any option more complicated than a plain call or put, with special rules about how or when it pays off. Some, for instance, only switch on if the price touches a certain level. Learn more →
Expected Shortfall (CVaR)Risk: The average loss you suffer on the worst days, the ones beyond your Value at Risk cutoff. It captures how ugly the tail really gets, which plain VaR ignores. Learn more →
Expected valueProbability: The long-run average outcome if you could repeat something many times, with each possibility weighted by how likely it is. It is the center of gravity of a random variable, not necessarily the most common result. Learn more →
ExpirationOptions: The deadline of an option. After this date an unused contract is worthless, like a coupon that expires, use it or lose it.
Exponential distributionDistributions: A distribution describing the waiting time until the next random event, such as how long until the next trade hits. It has no memory, meaning the wait ahead does not depend on how long you have already waited. Learn more →
ExposureTrading: How much of your money is at risk to a particular thing moving. If half your portfolio is in tech stocks, you have big exposure to tech.

F

FactorFactors: A common trait, like being cheap or recently rising, that helps explain why groups of investments earn different returns. Factors are the shared engines humming underneath many stocks at once.
Factor InvestingFactors: Deliberately tilting a portfolio toward traits that have historically paid off, such as value or momentum. Instead of picking individual winners, you bet on the trait itself. Learn more →
Factor Risk ModelFactors: A framework that breaks a portfolio's risk down into a few shared factors plus the leftovers unique to each holding. It helps you see what is really driving your ups and downs. Learn more →
Fama-French ModelFactors: An influential model that explains stock returns using a handful of factors like size and value, on top of the market itself. It showed that plain market risk was far from the whole story. Learn more →
Fat tailsStatistics: When rare, extreme outcomes happen far more often than a normal bell curve predicts. Financial markets have fat tails, which is why crashes and huge moves are more common than simple models expect.
FeatureMachine learning: In machine learning, an input you feed a model to help it make a prediction, like a stock's momentum or trading volume. Good features carry real information; useless ones just add noise.
Feature engineeringMachine learning: The craft of turning raw data into inputs a model can actually learn from, like converting prices into returns or smoothing out spikes. In finance this often matters more than the choice of model itself. Learn more →
Features and labelsMachine learning: Features are the inputs you give a model; labels are the correct answers you want it to learn to predict. The model's job is to figure out the connection between the two.
FillTrading: When your order actually gets completed. A full fill means your whole order went through; a partial fill means only some of it did.
Fixed IncomeInstruments: Investments that pay you a steady, predictable stream of money, mainly bonds. They are called fixed income because the payments are set in advance.
Forward contractDerivatives: A private agreement to buy or sell something at a set price on a future date. Like a futures contract, but customised and arranged directly between two parties rather than on an exchange.
FuturesInstruments: A contract to buy or sell something at a set price on a future date. Farmers, airlines, and traders use them to lock in prices ahead of time.
Futures contractDerivatives: A standardised agreement to buy or sell something at a set price on a future date, traded on an exchange. Both sides are locked in, and gains or losses are tallied up every day.
FX (Forex)Instruments: The market for trading one country's currency for another, like dollars for euros. It is the largest and most active market in the world.

G

GammaOptions: How fast an option's delta itself changes as the underlying moves. High gamma means the option's sensitivity can shift quickly, making it especially lively near the strike. Learn more →
GARCHVolatility: A statistical model that forecasts volatility by noting that calm and turbulent periods tend to cluster. After a big move, it expects more big moves to follow before things settle. Learn more →
GeneralizationMachine learning: A model's ability to perform well on new data it has never seen, not just the data it trained on. It is the entire point of machine learning, and the hardest part to get right in markets.
Gradient boostingMachine learning: Building a strong model by adding many small trees one at a time, where each new tree focuses on fixing the errors left by the ones before it. It is a workhorse method that often wins on messy, real-world data. Learn more →

H

HedgeTrading: A protective bet that offsets a risk you already have, like insurance for your investments. If your main position loses, the hedge softens the blow.
Hedge FundParticipants: A private investment firm for wealthy investors that uses aggressive strategies to chase high returns. They can bet on prices falling as well as rising.
HedgingRisk: Taking a second position designed to offset losses on something you already hold, a bit like buying insurance. It trims your risk but usually costs a little return.
HeteroskedasticityRegression: A fancy word for the spread of the errors not staying constant, for example being calm in quiet markets and wild in volatile ones. Ignoring it can make your confidence intervals and significance tests misleading. Learn more →
Hierarchical Risk ParityPortfolio: A newer portfolio method that first groups similar assets into a tree, then splits risk across the groups. It sidesteps the fragile math that trips up older optimizers. Learn more →
High-frequency tradingMicrostructure: Trading by computers that fire enormous numbers of orders in fractions of a second to capture tiny, fleeting edges. It lives and dies on speed and technology rather than any view on where prices head long term. Learn more →
HyperparameterMachine learning: A setting you choose before training that shapes how a model learns, like how deep a tree can grow. Tune too many by trial and error and you risk overfitting to your own test data.
Hypothesis testStatistics: A formal procedure for deciding whether the data give real support for a claim or could just be a fluke. You set up a boring default, then see whether the evidence is strong enough to overturn it. Learn more →

I

Idiosyncratic RiskRisk: The risk tied to one specific company or asset, like a bad earnings report or a factory fire. Because it is unique to that name, spreading across many holdings washes it out.
Implementation shortfallExecution: The gap between the price you saw when you decided to trade and the average price you actually ended up with. It bundles together spread, market impact, and delay into one honest tally of what execution really cost. Learn more →
Implied volatilityVolatility: The amount of future price movement that option prices are baking in. It is the market's forecast of turbulence, when option prices are high, implied volatility is high. Learn more →
In the moneyOptions: An option that would make you money if exercised right now. A call is in the money when the price is above the strike; a put is in the money when the price is below the strike.
In-sample vs out-of-sampleQuant research: In-sample is the historical data you used to build and tune a strategy; out-of-sample is fresh data you held back to test it honestly. A strategy that shines in-sample but flops out-of-sample was probably just memorizing noise.
IndependenceProbability: Two events are independent when knowing one tells you nothing about the other. Separate coin flips are independent; a stock and the overall market usually are not.
IndexMarkets: A scoreboard that tracks how a group of stocks is doing overall, like the S&P 500. It gives you a quick sense of whether the market is up or down.
Index FundInstruments: A fund that simply tries to copy an index, like owning a little of every company in the S&P 500. It is a cheap, hands-off way to invest in the whole market.
Information RatioPerformance: A measure of how much extra return a manager squeezes out relative to a benchmark, per unit of the risk they took betting against it. It grades skill at beating the index consistently. Learn more →
Informed vs noise tradersMicrostructure: Informed traders trade because they actually know something; noise traders trade for reasons unrelated to value, like rebalancing or hunches. Market makers earn from noise traders and lose to informed ones, so they price to survive the mix.
Intrinsic valueOptions: The part of an option's price you would pocket if you exercised it immediately. If a call lets you buy at $100 and the stock is $115, its intrinsic value is $15.
Inventory riskMicrostructure: The danger a market maker faces from holding a pile of stock that can drop in value before it is sold off. Managing this risk is why market makers adjust their quotes to lean away from positions they are stuck with.
IPO (Initial Public Offering)Markets: The first time a private company sells its shares to the public on a stock exchange. After the IPO, anyone can buy and sell the stock.
Iron condorStrategies: A combined position that profits when the price stays quiet inside a range. You sell options near the current price for income and buy further-out ones as protection.

K

Kelly CriterionRisk: A formula for how big to bet when you have an edge, aiming to grow your money fastest over the long run. Betting full Kelly is wild and stomach-churning, so many people deliberately bet a fraction of it. Learn more →
KurtosisStatistics: A measure of how heavy the tails of a distribution are compared with a normal bell curve. High kurtosis means extreme events happen more often than a bell curve would suggest.
Kyle's lambdaMicrostructure: A measure of how much the price moves for each unit of order flow, essentially the market's sensitivity to being pushed. A high lambda means the market is jumpy and illiquid, so trading size is expensive. Learn more →

L

LatencyMicrostructure: The tiny delay between when something happens in the market and when your system reacts to it. In fast trading, being a few microseconds slower can mean losing every good trade to someone quicker. Learn more →
Law of large numbersProbability: The more times you repeat a random experiment, the closer the average result gets to the true expected value. It is why casinos and insurers profit reliably over huge numbers of bets even though any single bet is uncertain. Learn more →
Ledoit-Wolf ShrinkagePortfolio: A popular, automatic recipe for cleaning up a covariance matrix so it behaves better in optimization. It pulls messy estimates toward a stable target by just the right amount. Learn more →
LeverageTrading: Using borrowed money to make a bigger bet than your own cash allows. It can multiply your gains, but it multiplies your losses just as much, so it is risky.
Limit OrderOrder types: An order to buy or sell only at a price you choose or better. You control the price, but the trade only happens if the market reaches your number. Learn more →
Linear regressionRegression: Fitting the best straight line (or flat plane) through your data to describe how an outcome moves with one or more inputs. It is the workhorse first model that quants reach for. Learn more →
LiquidityMarkets: How easily you can buy or sell something without moving its price. A big, popular stock is very liquid, you can trade lots of it instantly; a tiny obscure one is not.
Liquidity ProviderParticipants: Anyone who posts orders that others can trade against, making it easier for everyone to buy and sell. Market makers are the classic example.
Local volatilityVolatility: A model where volatility is not one fixed number but changes with the price level and the date. It is tuned to match all the option prices seen in the market today. Learn more →
Log Returns vs Simple ReturnsPerformance: Two ways of measuring percentage gains. Simple returns are the intuitive everyday version, while log returns add up neatly across time and are friendlier for math.
Lognormal distributionDistributions: The shape you get when the logarithm of a quantity is normally distributed. It only takes positive values and is skewed to the right, which makes it a natural fit for prices and other things that cannot go below zero. Learn more →
Long (Long Position)Trading: Owning something because you expect its price to go up. If you are long a stock, you make money when it rises.
Long an optionOptions: Owning an option that you bought. You paid the premium and now hold the right to exercise, and your downside is limited to what you paid.
Long/short equityQuant research: Owning stocks you expect to rise (long) while betting against stocks you expect to fall (short) at the same time. Done in balance it strips out the broad market and leaves your stock-picking skill on display.
Look-ahead biasQuant research: The mistake of accidentally using information in a backtest that would not have been known at the time, like today's closing price to make this morning's trade. It makes results look brilliant and is completely impossible to reproduce live. Learn more →
LotTrading: A standard bundle size for trading, such as 100 shares of a stock. Trading in lots keeps order sizes consistent across the market.
Low-Volatility AnomalyFactors: The surprising finding that calm, boring stocks have often delivered better risk-adjusted returns than wild ones. It upends the tidy idea that more risk always means more reward. Learn more →

M

Machine learningMachine learning: Teaching a computer to spot patterns from examples instead of programming every rule by hand. In finance it is powerful but tricky, because markets are noisy and the past does not repeat cleanly.
MarginTrading: Money you borrow from your broker to trade with more than you actually have. If the trade goes badly, you can be forced to add cash or sell, sometimes at the worst time.
Mark to MarketTrading: Updating the value of what you own to today's current price, even if you have not sold it. It shows your gain or loss as if you cashed out right now.
Market CapitalizationMarkets: The total value of a company on the stock market, found by multiplying its share price by how many shares exist. It is a quick way to size up how big a company is.
Market impactMicrostructure: The way your own order pushes the price against you as you trade, especially when trading big. Buy a lot and you nudge the price up; the bigger and faster you go, the more it costs. Learn more →
Market MakerParticipants: A firm that always stands ready to buy and sell a stock, quoting both a bid and an ask. They keep trading smooth and earn the small spread between the two prices. Learn more →
Market microstructureMicrostructure: The study of how trading actually happens up close: how orders meet, how prices form tick by tick, and who pays the spread. It is the plumbing beneath the price chart everyone else watches.
Market neutralQuant research: A portfolio built to profit whether the overall market rises or falls, by balancing bets that go up with bets that go down. The aim is to isolate your specific edge from the market's mood swings.
Market OrderOrder types: An order to buy or sell right now at whatever the current price is. It gets you in or out fast, but you do not control the exact price you get. Learn more →
Market-Cap WeightPortfolio: Sizing each holding by how large the company is, so the biggest firms take up the most room. Most well-known index funds are built this way. Learn more →
Maximum DrawdownRisk: The worst peak-to-trough fall a strategy suffered over a period, the deepest hole it ever dug. It is a gut-check for whether you could have stomached holding on. Learn more →
Maximum likelihood estimationStatistics: A go-to method for fitting a model by choosing the settings that make the data you actually saw as unsurprising as possible. In short, pick the story under which your observations look most expected. Learn more →
Mean (average)Statistics: Add up all the numbers and divide by how many there are. It is the everyday average, and it works well until a few extreme values drag it around.
Mean reversionQuant research: The idea that prices which have stretched unusually far from their normal level tend to snap back. Mean-reversion traders bet on the rubber band pulling the price back, which works until the trend keeps running. Learn more →
Mean-Variance OptimizationPortfolio: A method that picks portfolio weights to get the best trade-off between expected return and risk. It is elegant on paper but famously sensitive to small errors in its inputs. Learn more →
MedianStatistics: The middle value when you line all the numbers up in order, so half sit above and half below. Unlike the mean, it barely budges when a few outliers are wildly large or small.
Method of momentsStatistics: A simple way to fit a model by matching its theoretical averages and spreads to the ones you measure in the data. You line up the model's numbers with the sample's numbers and solve. Learn more →
Mid PriceTrading: The point exactly halfway between the bid and the ask. People use it as a fair estimate of what something is really worth right now.
Minimum Variance PortfolioPortfolio: The specific mix of assets that produces the calmest, least jumpy portfolio possible, ignoring return entirely. People reach for it when they care more about avoiding a rough ride than chasing gains.
ModeStatistics: The value that shows up most often in your data. A dataset can have one mode, several, or none if nothing repeats.
ModelMachine learning: A simplified math recipe that takes inputs and produces a prediction or decision, like guessing tomorrow's return from today's data. Every model is wrong in some way; the question is whether it is useful.
Modern Portfolio TheoryPortfolio: The foundational idea that you should judge an investment by what it does to your whole portfolio, not on its own. It shows how mixing assets can lower risk without giving up return. Learn more →
MomentumQuant research: The tendency for recent winners to keep winning and recent losers to keep losing, at least for a while. It is one of the most stubbornly persistent patterns in markets, though it can reverse sharply. Learn more →
Momentum FactorFactors: The tendency for recent winners to keep winning and recent losers to keep losing, at least for a while. It leans on the crowd's habit of piling into whatever is already moving. Learn more →
MoneynessOptions: A quick label for where an option stands versus the current price, in, at, or out of the money. It tells you at a glance how likely the option is to pay off.
Monte Carlo simulationProbability: Answering a hard probability question by having a computer play out the random scenario thousands of times and tallying what happens. When the math is too messy to solve directly, you just simulate and count.
MulticollinearityRegression: When two or more of a regression's inputs carry nearly the same information, the model cannot tell their effects apart. The predictions may still be fine, but the individual coefficients become shaky and hard to trust. Learn more →
Mutual FundParticipants: A pooled fund where many people's money is combined and managed together across many investments. It lets small investors own a diversified mix easily.

N

Neural networkMachine learning: A model loosely inspired by the brain, made of layers of simple units that together learn very complex patterns. Powerful but data-hungry and hard to interpret, which makes it risky on noisy financial data.
Normal distributionDistributions: The classic symmetric bell curve where most values cluster near the average and extremes are rare. It is the default assumption in much of statistics because so many things end up looking like it. Learn more →
NotionalTrading: The full face value that a trade controls, not the smaller amount of cash you put down. With leverage, a small deposit can control a large notional amount.
Null hypothesisStatistics: The boring default assumption that nothing special is going on, for example that a strategy has no real edge. A hypothesis test asks whether the data give you enough reason to reject it. Learn more →

O

Open interestOptions: The number of option contracts currently alive and unsettled in the market. High open interest means many traders hold that contract, a sign of activity and easy trading.
OptionOptions: A contract that gives you the right, but not the obligation, to buy or sell something at a set price by a set date. You pay a small fee up front for the chance to benefit if prices move your way, while your loss is capped at that fee.
Option greeksOptions: A set of nicknamed measures (delta, gamma, theta, vega, rho) that show how an option's price reacts to changes around it, price moves, time passing, volatility shifting, and rates changing. Learn more →
OrderOrder types: An instruction you give your broker to buy or sell something. It spells out what you want, how much, and sometimes at what price.
Order BookMarkets: The live list of all the buy and sell orders waiting at different prices for a stock. It shows where demand and supply are stacked up right now. Learn more →
Order flowMicrostructure: The stream of buy and sell orders hitting the market moment to moment. Reading which side is more aggressive gives clues about short-term price pressure, and it is prized information for fast traders.
Ordinary least squaresRegression: The standard way to fit a regression line by choosing the line that makes the total of the squared gaps between the line and the data points as small as possible. Squaring the gaps punishes big misses hardest. Learn more →
Out of the moneyOptions: An option that would be worthless if exercised right now, you simply would not use it. A call is out of the money when the price is below the strike; a put when the price is above it.
OutlierStatistics: A data point that sits far away from the rest of the pack. Outliers can be genuine surprises worth studying or just errors, and they can badly distort averages and models if ignored.
OverfittingRegression: When a model hugs the quirks and noise of the data it was trained on so tightly that it fails on anything new. It looks brilliant in the backtest and falls apart in live trading. Learn more →
Overfitting a backtestQuant research: Tuning a strategy so tightly to past data that it fits every random wiggle of history instead of a real, repeatable pattern. It looks perfect on the old data and falls apart on anything new. Learn more →

P

P-valueStatistics: The chance of seeing a result at least this striking if nothing real were going on. A small p-value (say under 5%) is the usual, but flawed, signal that the effect might be real. Learn more →
P&L (Profit and Loss)Trading: How much money you have made or lost on your trades. Green P&L means you are up; red means you are down.
Pairs tradingQuant research: Finding two stocks that usually move together, and when they drift apart, buying the laggard and shorting the leader on the bet they reconverge. It is the simplest, classic form of statistical arbitrage. Learn more →
PayoffOptions: What an option is worth at expiry, based only on where the price ends up. Drawn on a chart it makes the classic hockey-stick shape: flat while worthless, then rising once past the strike.
Penny StockInstruments: A very cheap stock, often under a few dollars, usually from a tiny or shaky company. They can jump around wildly and are easy to lose money on.
Poisson distributionDistributions: A distribution for counting how many times a rare event happens in a set window of time or space, like customer arrivals per hour. It works when events pop up independently at a steady average rate. Learn more →
Population vs sampleStatistics: The population is the entire group you care about; a sample is the smaller slice you actually measure. Statistics is largely the art of learning about the whole population from just a sample.
PortfolioTrading: The whole collection of investments you own, like stocks, bonds, and funds together. Spreading across a portfolio helps reduce the risk of any single one hurting you.
Portfolio CapacityPortfolio: The most money a strategy can manage before its own trading starts to spoil its returns. Small, clever strategies often break down once too much cash floods in. Learn more →
Portfolio WeightsPortfolio: The share of your total money sitting in each holding, usually written as percentages that add up to one hundred. Changing the weights changes the whole character of the portfolio.
PositionTrading: Any investment you currently hold in the market. Being in a position means you have money riding on how that thing moves.
Position SizingRisk: Deciding how much money to put into a single trade or holding. Getting size right often matters more to survival than being right about direction. Learn more →
PosteriorProbability: Your updated belief after combining your starting hunch (the prior) with fresh evidence. It is the answer Bayesian analysis is really after. Learn more →
PremiumOptions: The price you pay to buy an option, or collect when you sell one. Think of it as the cost of the coupon or the insurance policy itself.
PriorProbability: What you believed about something before seeing the latest data. In Bayesian thinking you always start with a prior and then let the evidence nudge it. Learn more →
ProbabilityProbability: A number between 0 and 1 that measures how likely something is: 0 means it never happens, 1 means it always happens, and 0.5 is a coin-flip. It is the basic language for talking about uncertainty. Learn more →
Probability density functionDistributions: A curve showing which values of a continuous random variable are more or less likely; taller regions are more probable. The area under any stretch of the curve gives the chance of landing in that range.
Probability distributionDistributions: A full description of all the values a random quantity can take and how likely each one is. It is the complete recipe for the randomness, from which averages, spreads and tail risks all follow. Learn more →
Prop TradingParticipants: When a firm trades with its own money to make a profit, rather than for clients. The firm keeps the gains but also eats the losses.
Protective putStrategies: Owning a stock and buying a put on it as downside insurance. If the stock crashes, the put gains value and cushions your loss.
Put optionOptions: The right (not the obligation) to SELL something later at a price fixed today. Like insurance that lets you sell a stock at $100 even if it crashes to $70, valuable when prices fall.
Put spreadStrategies: Buying one put and selling another at a lower strike. A cheaper, limited bet that the price falls, with both your cost and your payoff capped.
Put-call parityOptions: A fixed relationship linking the prices of a call and a put that share the same strike and expiry. If it ever breaks, traders can lock in a risk-free profit, so the market keeps it in line. Learn more →

Q

Quality FactorFactors: A tilt toward solid, profitable, well-run companies with steady earnings and low debt. The bet is that boring, dependable businesses quietly outperform the flashy ones. Learn more →
QuantQuant research: Short for quantitative analyst or quantitative trader. Someone who uses data, statistics, and code to find and trade patterns in markets, rather than relying on gut feel or a company story.
QuoteMicrostructure: A posted offer to buy or sell at a stated price and size. Market makers constantly post and update quotes, and the best buy and sell quotes together set the current spread.

R

R-squaredRegression: The share of the ups and downs in the outcome that your model manages to explain, from 0 to 1 (or 0% to 100%). Higher looks better, but a high R-squared can also be a sign you have overfit. Learn more →
Random forestMachine learning: A crowd of many decision trees, each slightly different, that vote on the answer. The crowd's average is far steadier and more accurate than any single tree, smoothing out individual mistakes. Learn more →
Random variableProbability: A number whose value is decided by chance, like tomorrow's stock return or the outcome of a dice roll. Instead of one fixed value it comes with a whole range of possibilities and how likely each one is. Learn more →
Realized volatilityVolatility: How much a price actually moved over some past stretch, measured after the fact. Comparing it with implied volatility tells you whether options turned out too cheap or too expensive. Also called historical volatility.
RebalancingTrading: Adjusting your portfolio back to your target mix after prices drift. If stocks grew to take up too much, you trim them and top up the rest to stay balanced.
RegimeQuant research: A stretch of time when markets behave in a distinct way, such as calm and rising, or panicked and volatile. Strategies that thrive in one regime can quietly break in another. Learn more →
Regression to the meanStatistics: The tendency for extreme results to be followed by more ordinary ones, simply because luck evens out. A star performer after a lucky streak, or a disaster fund after an awful year, both tend to drift back toward average.
RegularizationMachine learning: A set of techniques that deliberately keep a model simpler so it cannot chase every wiggle in the data. Trading a little accuracy on the training data buys much better behavior on new data. Learn more →
ReplicationDerivatives: Copying an option's payoff by trading the underlying and cash in just the right amounts over time. It is the core idea behind option pricing: if you can build the payoff yourself, you can price it.
ResidualRegression: The leftover gap between what actually happened and what your model predicted. Studying the pattern of residuals is one of the best ways to spot where a model is going wrong. Learn more →
RhoOptions: How much an option's price changes when interest rates move. Usually the least important greek for short-dated options. Learn more →
Risk of RuinRisk: The chance that a string of losses wipes out your account before your edge can pay off. Even a genuinely winning strategy can ruin you if you bet too big.
Risk ParityPortfolio: A way of building a portfolio so each holding contributes an equal share of the total risk, rather than an equal share of the money. It stops a few volatile assets from secretly dominating the whole thing. Learn more →
Risk-Adjusted ReturnPerformance: Return judged against the risk taken to earn it, rather than the headline number alone. A modest gain earned calmly can beat a big gain earned recklessly.
Risk-Free RateMarkets: The return you can earn with almost no risk, usually from safe government debt like Treasury bills. It is the baseline other investments are measured against.
Risk-neutral pricingDerivatives: A pricing shortcut where you value a derivative as if everyone ignored risk and just discounted expected payoffs at the safe interest rate. It is a mathematical trick, not a claim that investors truly do not care about risk. Learn more →

S

SABR modelVolatility: A popular model that describes how the volatility smile behaves, widely used in interest-rate and currency markets. It helps traders price and hedge options consistently across different strikes. Learn more →
Sample meanStatistics: The average calculated from your sample rather than the whole population. It is your best everyday guess of the true population average, though it wobbles a bit from sample to sample.
SamplingStatistics: The process of picking a subset of a larger group to study. Done well it gives a fair snapshot; done poorly, with bias in who gets picked, it quietly misleads you.
Sell-SideParticipants: The part of the industry that sells services to investors, like banks and brokers. They provide research, deals, and access to trade.
SettlementMarkets: The final step where the buyer's cash and the seller's shares actually change hands after a trade. It usually happens a day or two after you click buy.
Sharpe RatioPerformance: A score for how much return you earned for the amount of stomach-churning risk you took. Higher is better, and it lets you compare a wild strategy and a calm one on the same footing. Learn more →
Short (Short Selling)Trading: Betting that a price will fall. You borrow something, sell it now, and hope to buy it back cheaper later, pocketing the difference. It is riskier than going long because losses can pile up if the price rises instead.
ShrinkagePortfolio: A trick that nudges noisy, unreliable estimates toward a calmer, simpler average. In portfolios it keeps risk estimates from being fooled by random quirks in past data. Learn more →
SignalQuant research: A number or rule that hints at what a price might do next, like a stock being unusually cheap or its recent trend pointing up. A signal is only a nudge, not a guarantee, so quants blend many of them together. Learn more →
Signal-to-noiseQuant research: How much real, useful information a signal carries compared with random junk. Financial data is famously noisy, so a tiny bit of true signal buried in a mountain of noise is the everyday reality for quants.
Simpson's paradoxStatistics: When a trend that shows up in separate groups reverses once you lump the groups together, or the other way around. It is a warning that ignoring how data is split can flip your conclusion.
Size FactorFactors: The historical edge that smaller companies have shown over giant ones. Small firms are riskier and more overlooked, which can leave more room to grow.
SkewnessStatistics: A measure of lopsidedness in a distribution. Positive skew means a long tail stretching to the high side (a few big winners); negative skew means the long tail is on the low side.
SlippageTrading: The difference between the price you expected and the price you actually got. In fast or thin markets, prices can shift in the split second between placing an order and it filling. Learn more →
Smart BetaFactors: Index-style funds that follow a rule-based tilt toward factors like value or low volatility, rather than just tracking company size. It is a cheaper, systematic middle ground between plain indexing and active management.
Sortino RatioPerformance: A cousin of the Sharpe ratio that only counts downside swings as risk, ignoring the upside jumps nobody complains about. It rewards strategies that are volatile only in a good way. Learn more →
Spread (Bid-Ask Spread)Trading: The gap between the buy price and the sell price. A small spread means cheap, easy trading; a big spread means it costs you more to get in and out.
Standard deviationStatistics: The most common gauge of spread, and simply the square root of the variance so it comes back in the same units as the data. Roughly, it is the typical distance of a data point from the average.
Standard errorStatistics: How much a sample estimate, like the sample mean, would jump around if you redid the study with fresh samples. Bigger samples shrink the standard error, making your estimate steadier.
StationarityTime Series: A time series is stationary when its statistical behavior, like its average and its wobble, stays steady over time. Many models quietly assume stationarity, and prices famously break that assumption. Learn more →
Statistical arbitrageQuant research: Trading lots of small, related mispricings at once and relying on the law of averages rather than being right on any single bet. No one trade is a sure thing, but across hundreds the edge can show up reliably. Learn more →
Statistical powerStatistics: The chance a test will catch a real effect when one truly exists. More data and bigger effects give you more power and fewer missed discoveries. Learn more →
Statistical significanceStatistics: A result is called statistically significant when it would be unlikely to happen by pure chance under the boring default. It says an effect is probably not noise, but not how big or how important it is. Learn more →
Stochastic volatilityVolatility: A family of models where volatility itself is random and wanders over time, instead of staying constant. This captures how calm and stormy periods come and go. Learn more →
Stock (Share)Instruments: A small piece of ownership in a company. If you own a share, you own a tiny slice of that business and can benefit if it grows or pays out profits.
Stop LossRisk: A pre-set exit that sells a position once it falls to a chosen price, capping the damage. It takes the emotion out of cutting a loser.
Stop Order (Stop-Loss)Order types: An order that sits quietly until the price hits a level you set, then springs into action to buy or sell. People often use it to limit losses if a trade goes against them.
StraddleStrategies: Buying a call and a put at the same strike and expiry. You profit if the price makes a big move in either direction, a bet on turbulence, not on which way it goes.
StrangleStrategies: Like a straddle, but the call and put use different, wider strikes. Cheaper to set up, but the price has to move even more before you profit.
StrategyQuant research: A complete recipe for making money in markets: what to buy or sell, when, how much, and when to get out. In quant work a strategy is usually written as code so it can be tested on history and run automatically.
Strike priceOptions: The fixed price written into an option, the price at which you get to buy (for a call) or sell (for a put). Everything about the option is judged by how the current market price compares to this strike.
Supervised learningMachine learning: Training a model on examples where you already know the right answer, so it can learn to predict that answer for new cases. It is like studying with an answer key before the real exam.
Survivorship biasQuant research: Studying only the companies or funds that made it to today and quietly ignoring the ones that went bust or disappeared. This paints a rosier, safer picture than reality and can badly mislead a backtest. Learn more →
SwapDerivatives: A contract where two parties exchange streams of payments, for example, trading a floating interest rate for a fixed one. Handy for managing shifting costs or risks.
Systematic RiskRisk: The broad risk that hits the whole market at once, like a recession or a rate shock, which you cannot diversify away. Everyone feels it no matter how spread out they are.
Systematic tradingQuant research: Trading by a fixed set of rules that a computer follows the same way every time, instead of making one-off human judgment calls. The whole idea is to remove emotion and make decisions repeatable and testable.

T

T-testStatistics: A common test for whether two averages differ by more than you would expect from chance, like comparing a strategy's returns against zero. It is built for the small samples where the normal bell curve is a bit too optimistic.
Tail RiskRisk: The danger of rare but severe events, the once-in-a-decade crashes that standard models tend to underestimate. These extreme moves cause most of the real damage to portfolios.
Term structure of volatilityVolatility: How implied volatility differs across expiry dates, short-dated versus long-dated options. It shows whether the market expects turbulence soon or further off. Learn more →
Test dataMachine learning: Fresh examples held back and never shown during training, used to check whether the model actually learned something or just memorized. Peeking at test data while building the model defeats the whole purpose.
ThetaOptions: How much value an option loses each day simply from time passing, all else equal. It is the steady melting of time value as expiry nears. Learn more →
TickMarkets: The smallest amount a price is allowed to move, like one cent. Prices step up and down in these little increments.
Tick sizeMicrostructure: The smallest amount a price is allowed to move, like one cent. It sets the minimum bid-ask spread and quietly shapes how traders compete for the front of the queue.
TickerMarkets: The short code used to identify a stock, like AAPL for Apple or TSLA for Tesla. It is basically a nickname so traders can refer to it quickly.
Time seriesTime Series: Data recorded in time order, like daily prices or monthly sales, where the sequence itself matters. Because today often depends on yesterday, these need special tools that ordinary statistics lack.
Time valueOptions: The extra part of an option's price beyond its intrinsic value, what you pay for the chance that things improve before the deadline. It steadily melts away as expiry approaches. Also called extrinsic value.
Tracking ErrorPerformance: How much a portfolio's returns wander away from its benchmark. Low tracking error means you are hugging the index, while high means you are taking big independent bets.
Training dataMachine learning: The past examples a model learns from. If the training data is biased or unrepresentative, the model quietly inherits those flaws and carries them into every prediction.
Transaction CostsTrading: All the little costs of trading: broker fees, the bid-ask spread, and slippage. They quietly eat into your returns, so frequent trading can cost more than it looks. Learn more →
Trend followingQuant research: Betting that whatever has been going up will keep going up, and whatever has been falling will keep falling. It profits from big sustained moves and pays for that with many small losses when markets chop sideways. Learn more →
TurnoverQuant research: How much of a portfolio gets bought and sold over a period. High turnover means lots of trading, which racks up transaction costs and demands a strong edge just to break even.
TWAPExecution: Time-weighted average price, which slices a big order into equal pieces spread evenly over time. Trading steadily like a metronome helps hide your size and avoid slamming the price all at once. Learn more →
Type I errorStatistics: A false alarm: concluding there is a real effect when there actually is not one. In trading this is the classic sin of believing in a pattern that was really just luck. Learn more →
Type II errorStatistics: A missed discovery: failing to spot a real effect that was genuinely there. It is the quieter mistake, often caused by too little data. Learn more →

U

UnderfittingRegression: When a model is too simple to capture the real pattern, so it does poorly even on the data it was trained on. It is the opposite failure to overfitting. Learn more →
UnderlyingDerivatives: The actual thing a derivative is based on, a stock, index, currency, or commodity. The derivative's value simply rides on whatever the underlying does.
Uniform distributionDistributions: A distribution where every outcome in a range is equally likely, like a fair die or a well-made random number generator. There are no favored values, just an even spread. Learn more →
Unit rootTime Series: A technical property of a series that wanders off without settling back to any fixed level, like a random walk. If a series has a unit root it is not stationary, which changes how you are allowed to model it. Learn more →

V

Value at Risk (VaR)Risk: An estimate of the most you might lose on a normal bad day, given as a dollar or percentage figure at a chosen confidence level. It answers 'how bad is a typical bad day' but says nothing about true disasters. Learn more →
Value FactorFactors: The long-observed tendency for cheap, unloved stocks to beat expensive, glamorous ones over the long haul. You are betting that bargains eventually get recognized. Learn more →
VarianceStatistics: A measure of how spread out numbers are around their average. Small variance means the data huddles near the mean; large variance means it is scattered widely. Learn more →
Variance swapDerivatives: A contract that pays off based on how much a price actually swings around, letting traders bet directly on volatility itself. You profit if realized volatility comes in higher than the level agreed up front. Learn more →
VegaOptions: How much an option's price changes when expected volatility rises or falls. When markets turn jittery and volatility jumps, high-vega options gain value. Learn more →
VIXVolatility: A closely watched index that sums up the stock market's expected volatility over the next month, drawn from option prices. Nicknamed the fear gauge because it spikes when investors panic. Learn more →
VolatilityMarkets: How much and how fast a price swings up and down. High volatility means wild, jumpy moves; low volatility means calm, steady ones. Learn more →
Volatility (Risk)Risk: How much an investment's price swings up and down over time. In finance it is the everyday shorthand for risk, since bigger swings mean a bumpier, less predictable ride. Learn more →
Volatility skewVolatility: A lopsided version of the smile, where downside options (crash protection) cost more in volatility terms than upside ones. It reveals that investors pay up to guard against crashes. Learn more →
Volatility smileVolatility: A pattern where options far from the current price carry higher implied volatility than those near it, tracing a smile shape on a chart. It shows markets fear big moves more than a simple model expects. Learn more →
Volatility TargetingRisk: Adjusting how much you hold so your portfolio's swings stay near a chosen level, dialing exposure down when markets get wild. It aims for a steadier ride through both calm and stormy periods. Learn more →
VolumeMarkets: How many shares or contracts changed hands over a period, like a day. High volume means lots of activity and interest; low volume means it is quiet.
VWAPExecution: Volume-weighted average price, the average trading price over a period with bigger trades counting for more. It is a common benchmark for judging whether a large order was executed at a fair price. Learn more →

W

Walk-forwardQuant research: A testing method where you build the strategy on an early stretch of history, test it on the next stretch, then roll the window forward and repeat. It mimics real life, where you can only ever learn from the past and trade the future. Learn more →
Writing an optionOptions: Selling an option you did not previously own (also called shorting it). You collect the premium up front but take on the obligation to deliver if the buyer exercises, so your risk can be large.

Y

YieldInstruments: The income an investment pays you each year, shown as a percentage of its price. For example, a stock paying $2 a year while priced at $100 has a 2% yield.

Z

Z-scoreStatistics: How many standard deviations a value sits above or below the average. A z-score of +2 means the point is unusually high; it is a quick way to judge how extreme something is.