Get a ticker's history

Path parameters

Query parameters

Returns

Status codes

Notes

• Path history: GET /v2/series?tickers=NVDA,AMD (and its POST form) is the older spelling of this read. It keeps working, permanently, with the same response; its responses carry a Link header naming this path.
• Signals and bars come from different stores that don't always line up; a missing cell is null, never a dropped result, and _meta reports each column's source and each ticker's coverage.
• Names resolve by source, invisibly at the call site: every signal in the catalog reads its stored history, your own custom signals compile per result and behave exactly like built-ins, and open/high/low/close/volume/vwap/trades (or o/h/l/c/v/vw/n) read OHLCV bars.
• 1m / 1h / 1d are stored tiers. 1w is a real weekly resample: each signal samples the week's last observation, bars aggregate (open of first bar, max high, min low, close of last, summed volume and trades, volume-weighted vwap — null for a zero-volume week), results keyed by the week's Monday.
• 1q serves quarterly fundamentals on the fiscal grid: eps, eps_estimate, eps_surprise, eps_surprise_pct and every statement column 1y serves (below), under the same names and definitions, except the per-share ones (eps_basic, eps_diluted, shares_basic, shares_diluted; quarterly EPS is eps). Income and cash flow columns cover the quarter; balance sheet columns are at quarter end. Quarterly figures are as first reported, like 1y: a quarter a later filing restated keeps its original figures, and the fiscal fourth quarter is the annual report's year minus the first three quarters as first reported. Where a company restated an earlier quarter before its annual report, the four quarters do not add up to the annual figure (MSFT fiscal 2016 net_income: quarters 16,496M, annual 16,798M). Results are keyed by calendar quarter (2026-Q2) with each ticker's exact fiscal_period, reported (the date the quarter was announced, null when unknown), available_at (when the market could first trade on it, ISO 8601) and available_at_source (what decided it: report_time, filing_date or assumed). available_at is the axis a point-in-time backtest filters on, and what asof gates on: a post-market report is dated the report day but tradable only from the next open. All three come from the same report: the earnings release when the quarter has one, else the statement filing. The three revenue breakdown lists (revenue_by_segment, revenue_by_geography, revenue_by_product, below) are quarterly columns too. Quarterly columns cannot mix with daily ones in one request — the grids are not joinable.
• 1y serves annual financial statements, one result per fiscal year from fiscal 2010 on, keyed by the company's own fiscal-year label (t: "2025"; Walmart's fiscal 2026 ended on January 31, 2026). Each result carries fiscal_year, fiscal_period (the year-end date), reported (the filing date of the ORIGINAL 10-K, 20-F or 40-F, from SEC EDGAR — a later filing that repeats or restates the year never moves it) and available_at (9:30 AM ET of the first session after reported). Columns — income statement: revenue, cost_of_revenue, gross_profit, research_development, selling_general_administrative, depreciation_amortization, total_operating_expenses, operating_income, interest_income, interest_expense, total_other_income_expense, income_before_income_taxes, income_taxes, net_income (attributable to the company: the 10-K "Net income" line), net_income_consolidated (including noncontrolling interests), noncontrolling_interest_income, net_income_common (after preferred dividends), eps_basic, eps_diluted, shares_basic, shares_diluted (weighted averages), ebitda; balance sheet at year end: total_assets, total_current_assets, cash_and_equivalents, short_term_investments, receivables, inventories, property_plant_equipment_net, goodwill, intangible_assets_net, total_liabilities, total_current_liabilities, accounts_payable, debt_current, long_term_debt, total_equity, equity_attributable_to_parent, retained_earnings; cash flow: operating_cash_flow, investing_cash_flow, financing_cash_flow, capital_expenditures (negative, an outflow), dividends_paid (negative), change_in_cash, free_cash_flow (operating cash flow minus capital expenditures). Annual columns cannot mix with daily or quarterly ones in one request.
• At 1y: money is in US dollars; foreign filers' figures arrive already converted by the data vendor (at about the fiscal-year-end rate). The figures are as first reported: each year shows what its original annual report said, dated by that filing. Where a later filing restated a year, the original figures stay; a line the original report did not state on the same basis is null for that year (most often ebitda), never a later restated value. Per-share columns are on today's share basis (split-adjusted: AAPL fiscal 2010 eps_diluted is 0.54) and are null for ADRs, whose reports are per ordinary share, not per ADR (_meta.coverage says so). A company with several share classes has one EPS figure (Berkshire Hathaway: per Class B share). Banks' revenue is gross (interest income plus non-interest income), not the "total net revenue" banks headline. The last fiscal year's revenue_fy, net_income_fy, eps_diluted_fy and fy_period_end are also signals you can screen on.
• Revenue breakdowns at 1q and 1y: revenue_by_segment, revenue_by_geography and revenue_by_product are list columns, one list per fiscal period, read from the company's own SEC filings (10-K, 20-F, 40-F; 10-Q for quarters — the fourth quarter is inside the annual report and is not derived). Each element is {name, revenue, revenue_usd, share, currency, period_end, period_type, fiscal_year, fiscal_period}, the same element the signal of the same name carries for the latest period, in the filing's own order and with the company's own line names; a corporate or elimination line that brings the segments to the total carries reconciling: true and can be negative. A list is served only when its lines add up to the company's total revenue within 0.5%; otherwise the cell is null, never an estimate. revenue and share are in the reporting currency; revenue_usd converts at the average Federal Reserve H.10 rate over the period and is null for a currency H.10 does not publish. A fiscal period the statement columns lack still gets a result when a breakdown was filed for it (its scalar columns null). Under asof, a breakdown is visible from 9:30 AM ET of the first session after its filing date. _meta.sources names these filings. Filings from 2011 on.
• transitions_only=true filters results down to the state-change edges of the boolean signals — computed by the same shared rule that fires webhooks, so the two cannot disagree — with non-boolean signals riding along as context (the price at the flip). For a numeric threshold ("when did rsi_14 cross 70"), save it as a custom signal (rsi_14 > 70) and pass that with transitions_only.
• A company that changed its ticker reads its whole history under its current symbol: META from the 2012 IPO, FISV from 2003, with the earlier symbol named in prior_symbols on Tickers. A window that ends before the rename resolves to whoever held the symbol on its last day, the same way ?asof= does on the ticker read: META from mid-2021 until the rename is the Roundhill ETF that held the symbol then.
• At 1q, the two sources land at different times: eps arrives with the earnings announcement while the statement columns (revenue, net_income, operating_cash_flow, …) arrive with the filing. So the most recent quarter routinely carries an eps with revenue: null — a normal reporting lag, not a coverage gap. reported dates the announcement; available_at says when it could be traded.
• On a mixed state+bars daily request, spine entries on non-trading days (weekends/holidays) are dropped and counted in _meta.non_trading_days_dropped — they could never carry OHLCV and would read as gaps. transitions_only requests are exempt: their output is already sparse, so carry entries are never dropped and a flip recorded on one keeps its original date no matter which columns ride along (bar columns are null on such a result and count in _meta.coverage.*.missing_bar_cells).