Glossary: terms starting with R
60 terms from R² to Rule 10b5-1 plan, briefly explained and with a source where possible. All terms from A to Z
- R²
Coefficient of determination of a regression, between 0 and 1: the share of the variation explained by the explanatory variables (such as the market).
More: Beta and alpha: how closely a fund moves with the market
- Random matrix theory (RMT)
The theory of large random matrices, used in investing as a yardstick for noise in estimated correlation matrices: most of the spectrum of stock correlations looks like noise, and only a few large eigenvalues, such as the market factor, carry information. Methods that clean up the noise band of a matrix build on this.
Source: Laloux et al. (1999); Plerou et al. (1999); Bun et al. (2017)
More: Beyond Markowitz: estimation error, noise and true diversification · Methodology: Dimensions methodology: how Portfolio Lab counts independent risks
- Random walk (Random walk hypothesis)
Each change is independent of the previous ones. Prices in an efficient market should move roughly like this; deviations point to trends or mean reversion.
Source: Fama (1965); Lo and MacKinlay (1988)
More: The efficient market hypothesis: why it is so hard to beat the market
- Rank correlation (Spearman correlation, Spearman's rho)
The correlation of ranks rather than values (Spearman). It measures whether two variables rise together even if the relationship is not linear, and reacts less to outliers.
- Rational bubble (Rational speculative bubble, Rational expectations bubble)
A bubble that is consistent with rational behavior: investors buy despite overvaluation because the bubble grows fast enough to offset the risk of it bursting. The more likely the burst, the faster it has to grow.
Source: Blanchard and Watson (1982)
More: Speculative bubbles: what they are and how the LPPLS model detects them
- Real interest rate (Real rate, Real yield)
Interest rate after inflation. Exactly: (1 + nominal rate) / (1 + inflation) − 1; approximately: nominal rate minus inflation.
Example: 5% interest at 3% inflation: 1.05 / 1.03 − 1 = 1.94%.
Source: Fisher (1930)
- Real return
Return after inflation: (1 + nominal) / (1 + inflation) − 1.
Example: 7% nominal at 2.5% inflation is 4.39% in real terms.
- Realized volatility (Historical volatility)
Volatility that actually occurred, usually from the daily returns of the last 21 trading days, annualized with √252. On average it is below implied volatility.
More: VIX: how to read the stock market's fear gauge · Methodology: Volatility dashboard: methodology
- Rebalancing
Resetting the weights to the target allocation, either by trading or through targeted contributions.
- Rebalancing bonus (Diversification return, Rebalancing premium)
The extra return that regularly resetting to fixed weights can generate when assets fluctuate and are weakly correlated. It is not a guaranteed gain and can turn negative in long trends.
Source: Booth and Fama (1992); Willenbrock (2011)
- Recency bias (Recency effect)
Recent experience carries too much weight: after good years, investors expect too much, after crashes too little. That is why money often flows into funds after they have done well.
- Recession
A significant, broad-based decline in economic output lasting several months. As a rule of thumb, two quarters of falling gross domestic product count as a recession; in the US, a committee of the NBER officially dates the phases.
Source: Estrella and Mishkin (1998)
Methodology: Macro dashboard and regime analysis: methodology
- Recession probability (Recession risk, Recession probability model)
Probability, estimated from a model, that a recession will be under way in a future month. A well-known example is the New York Fed's probit model using the 10-year minus 3-month spread, twelve months ahead. It is a model figure, not a market price.
Example: With α = −0.6045 and β = −0.7374 and a spread of zero: Φ(−0.6045) = 27.3%.
Source: Estrella and Trubin (2006); Estrella and Mishkin (1998)
More: Market expectations: what the yield curve, inflation-linked bonds and options reveal about the future · Methodology: Macro dashboard and regime analysis: methodology
- Recovery rate (Recovery, Recovery ratio, Loss given default, LGD, Loss rate, Expected loss)
Share of a defaulted claim that the creditor ultimately gets back, for example by realizing collateral or through a restructuring. Its counterpart is the loss given default = 1 − recovery rate; the expected loss is the probability of default times the loss given default. It depends mainly on the seniority of the claim and on the collateral.
Example: Default rate of 5.0% over ten years and a recovery rate of 81.5%, as for project loans in Moody's data: expected loss of 0.93% over ten years.
Source: Moody's (2020)
More: Infrastructure investing: equity and debt, revenue models, valuation and research
- Redemption gate (Gate, Gating, Redemption limit)
A cap on redemptions per dealing date. If more investors want to exit, payouts are made pro rata; the rest waits for a later date.
Example: Limit $10 million, redemption requests $20 million: everyone receives 50% of their request.
- Reduction in yield (RIY, Cost drag, Effect of charges)
How much ongoing costs lower the annual return. If the costs are deducted from the fund's assets, (1 + r) × (1 − TER) − 1 remains. Over the full term, the fees also cost the return they would otherwise have earned.
Example: 7% before costs and a TER of 0.2%: 1.07 × 0.998 − 1 = 6.786%, i.e. 0.214 percentage points less per year.
More: ETF basics: costs, replication and distributions · Methodology: Methodology of the free calculators: FIRE, savings plan, withdrawal and leveraged ETF
- Reference portfolio
A simple, low-cost portfolio of index investments with the risk the owner is willing to bear, e.g. 80% equities and 20% bonds. It is the yardstick and the hurdle for the actual, actively managed portfolio.
Source: NZ Super Fund (2023); CPP Investments (2023); CalPERS (2025)
More: Strategic and tactical asset allocation and the total portfolio approach
- Reflexivity (Theory of reflexivity, Soros reflexivity, Boom-bust cycle)
Theory by George Soros: market participants' expectations do not just reflect reality, they also change it. Rising prices, for example, make financing easier and so improve the very figures on which the prices rest. In this way, excesses can reinforce themselves until they tip into a downturn.
Source: Soros (2003)
More: Legendary investors: the investment styles of the most famous investors, from Graham to Simons
- Regret aversion (Fear of regret)
The wish to avoid later regret. It explains why many people prefer to invest in installments: if the market falls after the first purchase, only part of the money is affected.
Source: Statman (1995)
- Regulated return (Regulatory asset base, RAB, Allowed return on equity, Incentive regulation, Network return, Grid return)
Return on invested capital that a regulator allows operators of natural monopolies such as electricity and gas grids. It is collected through network charges; the regulator weighs whether the grids remain financeable and whether consumers are protected from excessive charges. It is set for each regulatory period, so interest rate changes feed through with a delay.
Example: Draft by the Bundesnetzagentur (Federal Network Agency) for the gas grids 2028 to 2032: 40% equity at 5.76% before tax and 60% debt at 2.43% give a WACC of 3.76%.
Source: Bundesnetzagentur (2026)
More: Infrastructure investing: equity and debt, revenue models, valuation and research
- REIT (Real estate investment trust, REITs, Property stock, Mortgage REIT, Listed real estate)
Real estate investment trust: a listed property company that is exempt from tax at company level if it distributes most of its profit. Mortgage REITs hold mortgage loans instead of buildings.
Source: REIT Act (2007)
More: Real estate as an investment: rental property, REITs, open-end funds and private real estate · Methodology: Asset classes with a long history: methodology
- Relative momentum (Cross-sectional momentum, Relative strength)
Comparing several assets with one another: the strongest of recent months are bought and the weakest avoided (cross-sectional momentum).
Source: Jegadeesh and Titman (1993); Asness et al. (2013)
More: Tactical allocation: momentum, dual momentum and trend following
- Relative strength index (RSI)
An indicator by Welles Wilder: RSI = 100 − 100 / (1 + RS), where RS is the average gain over the last 14 periods divided by the average loss. Above 70 a price is considered overbought, below 30 oversold. This describes the most recent movement; it is not a forecast.
Example: Gains totaling 13.0 and losses of 3.0 points over 14 days: RS = 4.33, RSI = 81.25.
Source: Wilder (1978)
More: Technical analysis, chart patterns and candlesticks from a quant perspective
- Rental yield (Gross rental yield, Net rental yield, Purchase price factor, Rent multiplier, Price-to-rent ratio)
Annual rent relative to the price of a property. Gross: annual base rent divided by the purchase price. Net: rent minus the costs the landlord bears, such as management, maintenance and rent loss, divided by the purchase price plus acquisition costs. The price-to-rent ratio is the inverse of the gross rental yield.
Example: $300,000 purchase price, $30,000 acquisition costs, $12,000 base rent and $2,400 costs a year: gross 4.00%, net 2.91%, price-to-rent ratio 25.
More: Real estate as an investment: rental property, REITs, open-end funds and private real estate
- Repeat-sales index (Repeat sales, Repeat-sales regression, Repeat sales method, Mei Moses index)
Price index built from repeated sales of the same item: the price change between purchase and sale is spread across the periods in between by a regression. This sidesteps quality differences between items, but only uses items sold more than once, and these tend to come to market when their price has risen. Known from house price indices, and also used for art.
Source: Bailey et al. (1963); Mei and Moses (2002)
More: Collectibles as an investment: what art, wine, watches and classic cars really return
- Replication
How an ETF tracks its index: physically (all or selected securities) or synthetically via a swap.
- Replication crisis (Replication rate)
Dispute over how many published return patterns can be rebuilt. One large study found 65% of 452 anomalies insignificant; later studies arrived at replication rates of up to 82.4% depending on the measurement.
Source: Hou et al. (2020); Jensen et al. (2023)
- Repo (repurchase agreement) (Repurchase agreement, Sale and repurchase agreement, Haircut, Reverse repo)
The sale of a security with an agreement to buy it back at a fixed date and price. Economically, it is a secured loan. The haircut determines how much equity is required; via a reverse repo, bonds can also be obtained for short sales.
Example: $10 million of bonds with a 10% haircut: $9 million repo, $1 million equity, leverage 10.
More: Basis trades: positive and negative basis in bonds, CDS and futures
- Request for proposal (RFP, Asset management tender, RFI, Request for information)
Formal tender in which an institutional investor asks asset managers for a proposal for a mandate: questions on the firm, team, investment process, risk, performance, costs and operations, plus a fee proposal and references. It is often preceded by a shorter request for information (RFI), used to narrow the longlist down to a shortlist.
Source: CMA, Order (2019)
More: RFP and mandate awards: how institutional investors select asset managers
- Request for quote (RFQ, Request for quotation)
A request sent to several dealers at once for a tradable price for a given quantity; the trade is done with the best quote. The usual electronic route in bond trading and for large ETF orders.
Source: O'Hara and Zhou (2021)
More: Market making and bond trading: spread, inventory, axes, RFQ, cross trades and matrix pricing
- Resampled efficient frontier
Averages many efficient frontiers computed from slightly altered, simulated inputs. The averaged weights are more widely spread and fluctuate less than those of a single optimization.
Source: Michaud (1989)
- Residual
The part of a return that a regression model does not explain. The standard deviation of the residuals is the specific (idiosyncratic) risk of a fund or stock.
- Return contribution (Contribution to return, Contribution)
A position's share of a portfolio's total return: weight times return, chained over the periods. The contributions of all positions add up to the portfolio return.
Methodology: Backtest methodology: data, process, rebalancing and costs
- Return forecasts from valuation
High valuations have historically preceded lower returns over the following ten years. The predictive power is visible over long horizons, small in the short term and weaker out of sample.
Source: Campbell and Shiller (1998); Welch and Goyal (2008)
More: Shiller P/E (CAPE): how expensive is the stock market? · Methodology: Valuation with the Shiller P/E: methodology
- Return on equity (ROE, ROIC, Return on invested capital)
Net income divided by equity (ROE). Return on invested capital (ROIC), by contrast, relates operating profit after tax to all invested capital and can be compared directly with the WACC: only if it is higher does growth create value.
More: Fundamental analysis: valuing a company like an investor
- Return smoothing (Smoothing, Desmoothing, Unsmoothing, Appraisal smoothing)
Valuations of illiquid assets follow market prices with a lag and in muted form. As a result, volatility, correlation and beta appear too low, the Sharpe ratio too high, and reported returns are autocorrelated. Unsmoothing roughly reverses this.
Example: Smoothing parameter θ = 0.6: measured volatility = true volatility × √(0.4 / 1.6), i.e. half as high.
Source: Getmansky et al. (2004); Geltner (1993); Couts et al. (2024)
More: Private credit and public credit compared: returns, risks and what the research shows
- Return stacking (Return-stacked ETF, Stacked returns, Stacking)
Two sources of return on the same capital: futures or swaps are used to layer a second asset, such as bonds or managed futures, on top of a core such as equities. The second contributes only its return above the money market rate; losses from both sources add up.
Example: 100% equities (+8%) plus a 100% trend overlay (+6% with a 3% money market rate): 8 + (6 − 3) = 11%.
Source: Kung and Pohlman (2004)
More: Leveraged ETFs and return stacking: how they work, long holding periods, costs and research · Methodology: Backtest methodology: data, process, rebalancing and costs
- Return triangle
A table of annualized returns for every start and end year. It shows at a glance how strongly the result depends on the time of entry.
Methodology: Backtest methodology: data, process, rebalancing and costs
- Reversal effect (Reversal, Short-term reversal, Long-term reversal, Return reversal)
The observation that stocks with poor past returns later perform above average, and vice versa: in the short term from one month to the next, in the long term over three to five years. Strategies that regularly sell winners and buy losers have a tilt toward this effect.
Source: De Bondt and Thaler (1985); Jegadeesh (1990)
More: Equal weight vs cap weight: weighting, rebalancing and factor tilts
- Reverse convertible (Reverse convertible bond, Aktienanleihe)
A bond with a high coupon for which the issuer may deliver shares instead of the nominal value at maturity if the share closes below the strike. Decomposed: a bond plus a sold put; the coupon contains the put premium.
Example: Nominal value $1,000, strike $90: if the share closes at $70, you receive shares worth 1,000 × 70 / 90 = $777.78 plus the coupon.
- Reverse stress test (Reverse stress testing, Inverse stress test)
Stress test run backwards: the aim is to find the market movements that would cause a specified loss; for banks, failure; for investors, for example, a loss of 20% or missing a financial goal.
Source: Glasserman et al. (2015); Basel Committee on Banking Supervision (2018)
- Rho
Change in an option's price when the risk-free rate rises by one percentage point. Calls gain from higher rates, puts lose; for long maturities rho is noticeable, for short ones it is usually small.
More: Option pricing: where fair option prices come from · Methodology: Option pricer: how Portfolio Lab values options
- Risk budgeting (Risk budget)
The allocation is set in terms of risk shares rather than capital shares, e.g. 60% of the risk from equities. Risk parity is the special case of equal risk budgets.
Source: Roncalli (2013); Maillard et al. (2010)
More: Risk contribution: breaking portfolio risk down into its parts
- Risk contribution (Contribution to risk, Total risk contribution)
A position's share of portfolio volatility: weight times marginal risk contribution. The contributions of all positions add up exactly to the portfolio volatility.
Source: Roncalli (2013)
More: Risk contribution: breaking portfolio risk down into its parts · Methodology: Backtest methodology: data, process, rebalancing and costs
- Risk parity
Weighting in which each asset contributes the same amount to total risk. Low-volatility assets therefore receive more weight.
- Risk-adjusted return
Return relative to the risk taken. Metrics such as the Sharpe, Sortino, Treynor or Calmar ratio make portfolios with different levels of risk comparable.
- Risk-free rate
Return of a practically safe investment over the same period; in US dollars the Treasury bill yield, in euros the money market rate (€STR, formerly EONIA). It is subtracted from returns to measure excess returns.
- Risk-neutral probability (Risk-neutral density, Implied probability, Priced-in probability)
Probability read off prices when they are interpreted as expected values under the risk-neutral measure, for example from option prices. It contains the premium that investors pay for protection; for price declines, it is therefore usually higher than the actual frequency.
- Risk-neutral valuation (Risk-neutral measure)
Calculation technique in option pricing: you act as if all assets earned the risk-free rate on average and discount the expected payoff at that rate. The result equals the cost of replicating the option with the underlying and borrowing; it says nothing about real-world probabilities.
Source: Cox et al. (1979)
- Risk-on, risk-off
Market phases in which investors either crowd into risky assets together (risk-on) or flee to safe havens (risk-off). In risk-off phases, correlations between risky assets rise.
- RMW (Profitability factor)
Robust minus weak: return of profitable minus less profitable companies. Part of the five-factor model.
Source: Fama and French (2015); Novy-Marx (2013)
More: Factor investing: understanding the Fama-French factors · Methodology: Factor regression, factor allocation and factor statistics: methodology
- Robo-advisor (Robo advisor, Robo-advice, digital wealth management)
Digital wealth manager: after an online questionnaire, software assigns investors a risk level, invests the money in a model portfolio of ETFs, makes regular contributions and rebalances automatically. Costs: a fee as a percentage of assets plus the ongoing costs of the funds.
Source: D'Acunto et al. (2019); BaFin (2026)
- Robust z-score (MAD, Median absolute deviation, Modified z-score, Robust standardization)
Standardization using the median and the MAD (median absolute deviation from the median) instead of the mean and standard deviation: z = (x − median) / (1.4826 · MAD). The factor 1.4826 turns the MAD into the standard deviation for normally distributed data; individual outliers such as the pandemic months barely shift the median and the MAD.
Example: Median of the comparison values 1.0%, MAD 2.0 percentage points, current value −2.0%: z = (−2.0 − 1.0) / (1.4826 × 2.0) = −1.01.
Source: Rousseeuw and Croux (1993)
Methodology: Economic pulse: how Portfolio Lab calculates early indicators
- Roll critique (Roll's critique)
For every efficient portfolio, the linear relationship between beta and expected return holds exactly as a matter of math. The CAPM therefore only says that the market portfolio is efficient; because the true market portfolio cannot be observed, it cannot be tested directly.
Source: Roll (1977)
- Roll yield
Gain or loss as a futures contract converges toward the spot price at expiry. For VIX futures in contango, usually clearly negative.
- Roll-down (Rolldown, Roll-down return)
Price gain on a bond that, with an upward-sloping yield curve, moves over time into shorter maturities with lower yields. Together with the coupon, it makes up a bond's carry.
More: Yield curve: what the term structure reveals and why an inverted curve is seen as a warning sign · Methodology: Carry backtest: methodology for FX and bond carry
- Rolling (Roll, Futures roll)
Replacing an expiring future with a later one to keep the position open. Depending on the shape of the term structure, this costs or adds return; for VIX futures it usually means heavy losses.
Source: Whaley (2013)
More: VIX: how to read the stock market's fear gauge · Methodology: Volatility dashboard: methodology
- Rolling return
Return over a fixed window that moves forward month by month, for example every 10-year period of a backtest. Shows how strongly results depend on the starting point.
- Rolling volatility
Volatility over a moving window, e.g. 36 months. It shows that risk is not constant: it jumps in crises and falls in calm phases.
- Rule 10b5-1 plan (10b5-1 plan, Rule 10b5-1, Trading plan)
A trading plan fixed in advance that insiders in the US can rely on as a defense against insider trading allegations, provided they entered into it without knowledge of material non-public information. Since the 2022 amendments, directors and officers have been subject to a cooling-off period of at least 90 and at most 120 days.
Source: SEC, Rule 10b5-1 (2022)
More: Insider trading: what is prohibited, what is allowed and what insider purchases reveal
Brief, simplified explanations; not investment or tax advice.