Methodology

Sources and updates

Net asset values, scheme details and fund sizes come from the official disclosures of the mutual fund industry and the fund houses; fund portfolios from the fund houses' monthly disclosures and, for US ETFs, from the portfolio reports funds file with their regulator. Index values come from the index providers, prices from exchange and market data, and UCITS ETF names and share-class details from the European securities register and the issuers' own fund lists.

Prices, NAVs and returns are typically refreshed the next day. Every page shows the date its figures are as of.

Returns

• Indian mutual funds are measured on NAV, for the direct plan and growth option unless a page says otherwise, so dividends are not paid out but stay in the NAV.
• US and UCITS ETFs are measured on total return: closing prices with every distribution reinvested, in US dollars. A UCITS ETF's price in euros, pounds or pence is converted at that day's exchange rate.
• Trailing returns run from the last price on or before the start of the period to the latest price. Periods of a year or less are simple returns; three, five and ten years and since launch are annualised (CAGR), the yearly rate that compounds to the same result. A fund younger than the period shows none.

Rolling returns and SIP returns

A rolling return takes every window of the chosen length in a fund's history, a new one each day, and reports the median, the latest, the worst, the best and how many were negative. It shows how consistent a fund has been, which a single trailing figure cannot.

A SIP return assumes the same amount invested on the first of every month (or the next day with a price): ₹10,000 for Indian funds and $1,000 for ETFs, raised each January by any step-up you choose. Its return is the XIRR of those instalments against the value of the units bought.

Risk statistics

Risk is measured from five years of daily returns. Standard deviation is annualised over 252 trading days. Sharpe and Sortino ratios measure return above a risk-free rate, 6.5% a year in rupees and 4% in dollars, per unit of total and of downside volatility. Maximum drawdown is the largest fall from a peak.

Beta, alpha, R², upside and downside capture, tracking error and the information ratio compare a fund with its benchmark where a benchmark series covers the fund's whole history, and with its category average otherwise. Each page says which. A statistic needs at least 60 daily returns and is left blank below that.

Benchmarks and category averages

A fund's benchmark is the one its fund house names. Its daily values come from the index provider's own total return index where one is published; otherwise from the longest-running fund that tracks the index, an overseas ETF converted to rupees, or a blend of indices built from the benchmark's stated weights. The source is shown in brackets wherever it is not the index itself, because a tracking fund runs slightly below its index after costs.

A category average is an equal-weighted index of every fund in the category, starting at 100 on the first day any of them reports. Days on which fewer than half the members report are skipped, and a fund whose price history has a break is left out of the average.

Assets and costs

• AUM for Indian funds is the average assets under management for the latest quarter, summed across a scheme's plans. The totals on category and fund-house pages are our sum over the funds listed, and can differ from a fund house's own published figure, which usually counts month-end assets.
• Assets for US ETFs are the latest reported total net assets. UCITS sizes are as each issuer reports them: some give the share class alone, most the whole fund across its classes, and each page says which.
• Expense ratios are the total expense ratio as reported, for the plan or share class shown.

Corrections to source data

When a fund re-bases its units, for example a 1-for-10 split, the history before the change is adjusted so returns run straight through it, while today's price stays exactly as published. Price data occasionally carries errors of its own: a unit change left unadjusted, a single bad print, a stretch quoted in the wrong currency or in pounds instead of pence. These are corrected only where the evidence is unambiguous, for a currency stretch only when the fund's own NAV confirms which side is right, and left as published otherwise. A gap is preferred to a guessed number.

Portfolio overlap

The overlap between two funds is the sum, over every holding they share, of the smaller of its two weights. Two funds holding the same stocks in the same proportions overlap 100%. Cash, receivables and derivatives are left out, and holdings are matched by name for Indian funds and by security identifier for US ETFs.

Momentum rankings

Returns over one day to one year are ranked across the list in view, 1 being the best. The combined rank ranks the sum of each stock's 3-month, 6-month and 1-year ranks. The z-score averages how far each of those three returns, taken as log returns, sits above or below the list's mean, in standard deviations. Relative strength compares a stock's change over 55 trading days with its index's. Picking a universe, sector or industry ranks again within it.

Portfolio tracker

Gains are worked out lot by lot, first in, first out. A holding's return is annualised as XIRR once it is a year old, and shown as a plain return before that, because annualising a few weeks magnifies noise. Tax figures follow Indian capital gains rules and are estimates for planning, not a tax computation. All of it runs in your browser; see the privacy policy.

Limits

Figures can be delayed or wrong when a source is, and some funds lack a figure their source does not publish. Past returns do not predict future ones. Check anything you act on against the fund's own documents. If something looks wrong, tell us.