ICICI Prudential Innovation Fund

Active share (Cremers & Petajisto, 2009) is half the sum of the absolute difference between this fund's disclosed weight and the benchmark proxy's disclosed weight, security by security. Tracking error (Roll, 1992) is the annualized standard deviation of the daily difference between this fund's return and the proxy's, over the days both have a stored NAV. Neither is a rating or a score; both describe how much this fund's actual holdings and returns differ from a portfolio that simply held the benchmark.

StockProof has no official total-return index (TRI) data — the price feed it stores tracks an index's PRICE only, which understates a real TRI benchmark by roughly its dividend yield. Instead, both figures are computed against another fund: an index fund that tracks the same benchmark, whose own NAV already reflects reinvested dividends and whose own disclosed portfolio stands in for the benchmark's constituent weights. Measured directly: two different index funds on the same benchmark differ from each other by only 2–3 basis points a year of tracking error — two orders of magnitude below any actively managed fund's real figure — which is what licenses using one as a stand-in for the true benchmark. This proxy is hand-picked and verified for 24 declared benchmarks; a fund benchmarked to anything else shows no active share or tracking error, not a guessed one.

Tracking error is computed over whatever NAV history both funds actually share, which is genuinely short — AMFI's history file starts no earlier than April 2006, but the DEPTH stored here reaches back only to around September 2023, roughly three years, well short of the five-year-plus history stock pages elsewhere on this site commonly show. The exact window and day count are shown beside every figure so it is never mistaken for a longer, more settled estimate.

Higher active share means this fund's holdings differ more from the benchmark proxy; lower means they resemble it more closely. Neither direction is automatically better or worse — it depends on what the fund is being held for and whether that difference is deliberate stock-picking or an accident of the universe boundary this data can see. The same is true of tracking error: a fund built to closely track its benchmark should show a low one, while a genuinely active fund is expected to show a higher one by design. This is descriptive, not a recommendation to buy, sell or hold anything, and nothing on this page ranks or scores funds against each other.

A single day whose NAV changes by an implausible amount — almost always an unrecorded corporate action such as a unit split, not a real market move — is detected and the whole tracking-error figure is withheld for that fund rather than shown corrupted by one bad day. Active share is unaffected by this check: it comes from the fund houses' own disclosed portfolios, not from NAV history.