Portfolio glossary

• Portfolio Builder: the three-stage tool where a portfolio is configured and built—Portfolio Setup, Asset Selection, and Asset Allocation. See Setting up a portfolio.
• Backtest: the simulation of a portfolio's strategy over its full history. Building a portfolio runs the backtest; everything you see in the summary and in Insights is its result.
• Asset universe: the set of assets a portfolio may choose from, built from asset lists and narrowed by filters. The portfolio holds a selection from the universe, not the whole universe. See Selecting assets for a portfolio.
• Performance Portfolio: the part of the portfolio built to generate returns. Every portfolio has one.
• Hedge Portfolio: the optional second part, built to offset the Performance Portfolio's swings by selecting assets that move independently of it, or against it. See Performance and Hedge portfolios.
• Benchmark: the asset a portfolio is charted against in the portfolio summary and in Insights. A viewing choice, not part of the portfolio's configuration—any asset on the platform can serve.
• My Portfolio: the one portfolio you mark as the portfolio you actually follow; pfolio keeps its allocations in a dedicated asset list and uses it as the default in the rebalancing tool. See Managing your portfolios.

• Rebalance: the scheduled rebuild of the portfolio from the latest data—assets re-selected, weights re-set. The whole pass is walked through in How a portfolio rebalances.
• Rebalance frequency, Period, and Start Point: how often the rebalance happens (weekly, monthly, or quarterly), every how many of those, and where in each period the rebalance day falls. See Rebalance frequency.
• Rebalancing period: the stretch between two rebalances, over which one allocation is held unchanged.
• Allocation: the set of weights the portfolio holds—which assets, and how much of each. The old allocation applies through the close of the rebalance day; the new one applies from the next trading day's open.
• Turnover: how much trading the switch from one allocation to the next requires. More frequent or more reactive rebalancing means more turnover. See Controlling turnover and adaptation speed.
• Transaction costs: the estimated cost of that trading, configured per portfolio and optionally deducted from the backtest's returns. See Transaction costs.

• Goal: a weighted set of metrics that ranks the universe at each rebalance; the top-ranked assets become the period's selection. See Asset allocation goals.
• Goal metric: one metric within a goal, with its own weight and its own lookback—Sharpe ratio measured over 250 days, say.
• Lookback window: how much history a metric or estimator sees. Short windows react quickly but are noisy; long windows are stable but slow to adapt. See Lookback windows.
• Ranking inertia: a bonus that protects current holdings in the ranking, so they are not rotated out by marginally better newcomers. See Controlling turnover and adaptation speed.
• Allocation inertia: the same idea applied to weights—the new weights lean towards the old ones unless the data argues otherwise.
• Constraints: the bounds the optimiser must respect—the per-asset allocation range, the cash range, and the split between the Performance and Hedge portfolios. See Asset allocation constraints.
• Cash allocation: the share of the portfolio held as cash rather than invested, kept within the range set in the constraints.
• Leverage: scaling the portfolio's exposure beyond 100% of its value. Set in the Portfolio Setup advanced settings.

• Optimisation method: the algorithm that turns the period's selection into weights. Six are available; the overview and how to choose are in Portfolio optimisation.
• Markowitz mean-variance methods: the family of four methods (Risk Level, Efficient Risk, Efficient Return, Max Quadratic Utility) that trade expected return against risk, each optimising a different target. See Markowitz mean-variance methods.
• Hierarchical Risk Parity (HRP): the method that clusters assets by correlation and allocates risk down the hierarchy—no expected returns needed. See Hierarchical risk parity.
• Equal Weight: the same weight to every selected asset; needs no estimates at all. The transparent baseline any other method should beat.
• Expected-return estimator: how the Markowitz methods estimate what each asset will return. Often matters more than the choice of method itself.
• Risk (covariance) estimator: how a method estimates the assets' volatilities and co-movements—the input every method except Equal Weight relies on.

• Drawdown: the decline from a previous peak—what the portfolio lost before recovering. Often the better view of how a portfolio would have felt to hold.
• Contribution: a holding's share of the portfolio's gains or losses, shown in the contribution charts. See Understanding the allocation charts.
• Group By: the aggregation control on the allocation charts—it sums assets of the same category (asset class, region, and so on) into a single series.

The general metric definitions—Sharpe ratio, volatility, and the rest—live in the metrics we use.