The Crypto Times India USDC Premium Index · daily fix 11:30–12:30 IST · CTPI-USDC-SPOT-INR

Methodology v0.3.0 · effective 2026-07-30. Prior versions: see the changelog.

The daily fix is computed from order-book snapshots captured in the 11:30–12:30 IST window. Each constituent venue’s price is the median of its mid-quotes in the window, weighted by the median two-sided INR depth resting within ±100bps of mid. The index value is the depth-weighted median across eligible venues — moving it requires real resting depth, not printed prices. With two constituents, a depth-weighted median resolves to the deeper book’s price; on most days that is Coinbase. The sister index CTPI-USDT uses an unweighted median across a wider venue set; CTPI-USDC weights by depth because a two-venue universe makes an unweighted median vulnerable to a single thin book.

Constituents. Coinbase (INR order books) and KoinBX. Giottus is collected but excluded while its USDC book is inactive. Constituents are reviewed against the eligibility rules below; changes are versioned in the changelog.

Eligibility per window. An uncrossed book with at least ₹20,000 of two-sided depth within the band and at least 2 snapshots. The ₹20,000 floor is sized to current India-facing USDC book depth and will be revisited as venues grow. A snapshot minimum is used instead of a coverage floor because, with two venues, a coverage floor would routinely reduce the fix to a single book; depth weighting bounds the influence of thin participation instead. Venues failing structural quality review — for example, a book that does not update — are excluded at constituent level and disclosed.

References. The raw premium compares the fix to the ECB-derived USD/INR cross (EUR/INR ÷ EUR/USD, previous ECB business day). Around ECB holidays the most recent rate is carried for up to 5 calendar days; beyond that no daily fix is published for the day (fx_missing). The peg-adjusted premium additionally divides out the global USDC/USD rate (Kraken spot, single-sourced for its consistently deep USDC/USD book); if it is unavailable in-window, the fix publishes the raw premium with a data-quality flag and omits the peg-adjusted figure.

Weekends and holidays. No daily fix is published; an indicative reading is computed at the same time against the last ECB business day’s USD/INR, labelled with that rate’s date, marked indicative in the series and downloads, shown as a gold point on the chart, and never counted in the daily fix statistics.

Publication and failure. Daily fixes require two or more eligible venues and publishable references and are immutable once written. If fewer than two venues pass eligibility, no fix is published for that day — the gap stays visible in the series and the cause is recorded in the engine’s run log. Venue price disagreement above 150bps is disclosed on the fix as high venue dispersion, never hidden.

Live tracker and staleness. The live block is an indicative reading, sampled roughly every 60 seconds during Indian market hours and every 5 minutes overnight, computed with the same depth-weighted aggregation as the daily fix. A sample counts only if it passes the same admission rules the fix applies — a successful capture, an uncrossed book, no disqualifying quality flag, a mid, and non-zero depth inside the ±100bps band — so a venue that keeps answering with an empty book cannot keep the tracker looking live. If the newest qualifying sample is more than 15 minutes old, the live block stops showing a figure and shows a STALE label with the age of the last sample. Fifteen minutes is fifteen missed samples at the peak cadence and three overnight: long enough never to fire on one slow pass, short enough that no reader can mistake an hours-old number for a current one. The daily fix is computed from the closed 11:30–12:30 IST window, is a separate measurement, and is published and displayed normally while the tracker is stale. Effective 2026-08-31; no published value changes.

Publication times. The daily fix is published at 12:31 IST on IST business days, after the 11:30–12:30 window closes. Missed or late publication. If the scheduled computation does not run (host or scheduler failure), the fix job runs as soon as the engine recovers, from the samples captured inside the window and the same previous-business-day USD/INR reference, and produces the same value an on-time run would have. If no eligible samples were captured inside the window, no fix is published for that date — ever — because a fix is never estimated from samples taken outside its window. Either case is recorded in the corrections log as a publication-timing entry, not a revision. Mumbai exchange holidays are not currently treated as non-trading days: the venues trade continuously, so a fix is computed on any Monday to Friday. If that changes it will be a versioned methodology change, not a silent one.

USDC−USDT spread sign. spread_pp is the USDC peg-adjusted premium minus the USDT peg-adjusted premium. A negative spread_pp means USDC is the cheaper dollar; a positive spread_pp means USDT is. Both indices’ per-₹1,00,000 outcomes are published alongside it, so a comparison between the two is a subtraction of two stated figures rather than a re-derivation at other rates. Effective 2026-08-31; the convention is unchanged, it was previously undocumented.

Rounding and precision. The global USDC/USD rate is published to 6 decimal places together with the formula that consumes it, and the unrounded value captured from the venue is retained in the stored fix and in the API payload. The index value is published to 4 decimal places; premia to 2 for display and 3 in the API.

Restatement. A restatement is dry-run reviewed before being applied and recorded permanently in the corrections log with the original value, the corrected value, the reason, and both timestamps; restated fixes are annotated in the corrections log alongside the series.

Statistics. Rolling range statistics use 30 calendar days of the published series and include the marked pilot period (27–29 Jul); pilot fixes were computed under pre-release parameters, are flagged in every download, and are included so the statistic reflects the full published record.

2026-09-02 · operational (no methodology version change) — Live reading: the raw and peg-adjusted premiums now divide by the same USD/INR reference — the rate in force when the sample was taken, which is the one shown beside the reading; until now the peg-adjusted figure used the latest rate and the two could differ across an ECB update, which the reading disclosed with a denominator_shift flag. That flag is retired. Per-exchange table: a quote’s staleness is judged against the sampling interval in force when it was taken, so the 08:00 IST switch from overnight to daytime cadence no longer marks every venue stale for one cycle. The payload now carries the citation line, the disclaimer and the risk note, so the index pages print the engine’s own words; the index name in those lines is now written The Crypto Times India USDC Premium Index (family naming decision). The methodology states what happens when a scheduled fix is missed or late; the corrections log records the missed fix of 2 September 2026. Housekeeping: HEAD requests no longer leave a connection in body-less mode; the shipped configuration names the production CTPI-USDT feed so a fresh install shows the comparison without a manual step. No published index value, no methodology parameter and no historical figure changes in this release.
2026-08-31 · operational (no methodology version change) — Live tracker: a published 15-minute staleness threshold added; past it the live block shows a STALE label with the age of the last sample instead of a figure, and the daily fix continues to display unaffected. Staleness and the /v1/health verdict are now judged on the last qualifying capture rather than on the last write of any kind, which could report healthy while the live block had nothing to draw. A /status endpoint returns a non-zero HTTP status while stale so the condition can be alerted on. Live readings now publish the USD/INR denominator that produced them, alongside the percentage, the rate and the sample time. The peg-adjusted premium becomes the headline figure on the page, the citation line and the comparison panel, with the raw premium shown directly beneath and labelled; the global USDC/USD rate is shown to 6 decimal places with its formula. The spread_pp sign convention is now documented. “Today’s range” states the window it actually covers and the number of samples in it, correcting copy that described it as 24 hours. A “next fix” line is shown, driven by the same business-day calendar the fix engine uses. No published index value, no methodology parameter and no historical figure changes in this release.
2026-08-31 · internal — Snapshot admission rules unified. The live tracker had required a completely empty quality-flag field while the daily fix and the venue table rejected only the two disqualifying flags, so a benign informational flag could empty the live tracker while every other surface reported healthy. All three now share one definition. Published fixes are unaffected: no historical fix used a snapshot that this change would have admitted or excluded differently.
2026-07-30 · methodology v0.3.0 — Fix window aligned to the family standard shared with CTPI-USDT: 11:30–12:30 IST. The pilot fixes of 27–29 Jul were computed under pre-release parameter sets (12:30–12:40 window) and are marked pilot in the series and downloads.
2026-07-29 · methodology v0.2.0 — Depth band set to ±100bps (a ±50bps band under-measured wide-spread books); per-instrument depth floors calibrated during the pilot period, to be re-reviewed at 30 days; venue dispersion disclosure added. Constituent review: KoinBX retained despite persistent positive deviation — its USDC/INR quotes reflect genuine retail-accessible prices for that instrument (its USDT book is assessed separately under CTPI-USDT), its influence is bounded by depth weighting, and dispersion is disclosed on every fix.
2026-07-27 · methodology v0.1.0 — Series begins. Collection live across three India-facing venues; one venue’s USDC book excluded for inactivity.

Any published fix that is subsequently revised, any material error in commentary, and any scheduled fix that is not published, is recorded here permanently. Daily fixes are immutable once written; if a restatement is ever required it follows the restatement rule in the methodology — dry-run reviewed, then applied, then recorded here with the original value, the corrected value, the reason, and both timestamps.

Publication timing — 2026-09-02: no daily fix

No daily fix exists for Wednesday 2 September 2026. The index host was powered off for a plan resize in the morning IST (the operator reports about 06:30) and was not powered on again until 16:40 IST, so nothing was sampled during the 11:30–12:30 IST observation window. On boot the persistent fix timer ran the scheduled job immediately; the engine found no eligible samples in the window and refused to publish (insufficient_venues(0<2)), which is the behaviour the methodology requires: a fix is computed only from samples taken inside the window, and samples that were never taken cannot be recreated or estimated.

Collection resumed at 16:44 IST and the live tracker was blind until then. The 1 September fix remained the latest official value and every surface showed it with its own date. No published value was created, revised or withdrawn — this is a gap in the series, disclosed, not a revision. Detected by internal review of the boot log the same day.

The daily fix series to date. Full series JSON with a stable schema: /v1/fixes.json.
CSV JSON

This page can’t answer that for you — it can only price the choice. The decision panel above shows today’s spread — the excess premium USDC carries over USDT — and what it costs per ₹1,00,000. What you get for that premium is a different issuer and reserve structure. Both numbers update every trading day.

The same frictions that create the USDT premium — capital controls, tax treatment, on-ramp costs — apply to USDC, with thinner order books and fewer venues on top. Scarcity of supply on India-facing books has generally priced USDC slightly above USDT in this series, though the spread has briefly inverted.

Coinbase (INR order books) and KoinBX are current constituents. Giottus is collected but excluded while its USDC book is inactive. Constituents are reviewed against published eligibility rules; changes appear in the changelog.

No. These are public, on-exchange order-book prices that any KYC-verified Indian user can access. The premium is a market phenomenon, not an unofficial exchange rate.

Crypto trades every day but the ECB reference doesn’t update. Weekend and holiday readings are published as indicative, measured against the last business day’s USD/INR, and marked distinctly on the chart.

The per-exchange table shows sampled order-book best bid and ask — tradeable quotes at sample time, subject to depth and fees. The index value itself is a computed benchmark, not a quote.

Nearly, but not exactly — USDC trades within a few basis points of $1 globally; the current rate is shown in the metadata line above. The peg-adjusted premium divides this out so the India-specific friction is isolated.

The citable number is the once-daily fix. A live indicative reading updates continuously above it and is never part of the daily fix series.