1\section{Data}2\label{sec:data}34\subsection{Source and license}56All market and accounting data come from FinData OS, a private lakehouse of FirstRate Data operated by the author and7accessed through its versioned API (\texttt{findata.spboucher.ai/api/v1}). The lake holds 6.7 billion raw 1-minute8bars for 16,195 U.S. stock symbols (2000--2026, including 8,069 delisted symbols), quarterly fundamentals for9\nEntities{} entities (67 metrics, filing dates 2012--2026) and 373,326 earnings-release events. Bars are timestamped10at the bar open in UTC and cover 04:00--20:00 Eastern; zero-volume minutes are absent at the source and are never11imputed. FirstRate Data is licensed for private research use; no raw record is redistributed, and the replication12package contains code, hashed extraction manifests (dataset identifiers and versions, row counts, SHA-256 digests) and13derived aggregates only. Public augmentations are the SEC's \texttt{company\_tickers.json} and \texttt{submissions}14files (CIK, SIC industry), and the Kenneth French daily factor library.1516\subsection{Universe and survivorship}1718The accounting panel contains the \nEntities{} entities for which the vendor supplies quarterly statements; it has19\emph{no delisted entity}, because the vendor ingestion covered tickers active on 1 November 2026. The price side is20survivorship-free (delisted names are present in the bar lake), but the firm-quarter sample is conditioned on survival21to the end of the sample. This is a structural caveat: firms that deteriorated and delisted are absent, so adverse22base rates are understated and the most vulnerable firms are under-represented. We state it in the abstract and23treat it as the first limitation. Table~\ref{tab:universe} reports the construction of the universe: \nEligible{}24firms with at least eight quarterly periods, at least 250 sessions of 1-minute bars after 2012 and a non-governmental25SIC division; \nWithCik{} are matched to a CIK. Identifiers are vendor tickers; the vendor merges ticker changes under26the latest ticker, and we audit reuse with the SEC's former-name history.2728\subsection{Reveal dates and point-in-time discipline}2930Earnings-release dates are matched to fiscal quarters as the first release in $[E_q, F_q+10\text{ days}]$;31\releaseMatchedShare\% of firm-quarters are matched with a filing lag of at most 30 days (median lag between release32and filing: \filingLagMedian{} day). Release timing is known (before-open or after-close) for about two thirds of events.33Structured facts carry \texttt{filing\_date}, \texttt{available\_at} ($=$ filing date $+1$ day, a documented policy34rather than a historical vintage) and a revision identifier; because every fact was first ingested in November 2026 and35carries revision~1, restatements that occurred before ingestion are not reconstructible. Inputs therefore use prior36quarters only, outcomes use the vendor's current value of quarter $q$, and the stringent as-originally-reported37specification is left for a future SEC-filing-level reconstruction (Section~\ref{sec:limitations}).3839\subsection{Sample}4041The quarter panel has \nFirmQuarters{} firm-quarters for \nFirmsPanel{} firms. After the admissibility and exclusion42cascade of Table~\ref{tab:cascade} (market window of at least 15 valid sessions, raw close of at least one dollar,43available fundamentals, prediction of the next quarter, defined expectation, regular fiscal calendar), the event-time44sample has \nSampleEvent{} observations across the five horizons (\nSampleFirms{} firms, \nSampleQuarters{} firm-quarters)45and the primary $h=20$ cohort has \nObsPrimary{} observations with cutoffs from \firstCutoff{} to \lastCutoff; the46fixed-calendar sample has \nSampleCalendar{} firm-months. The adverse outcome occurs in \baseRatePrimary\% of47primary-cohort firm-quarters. Table~\ref{tab:summary} reports summary statistics and Figure~\ref{fig:coverage} the48coverage by year together with the adverse rate.4950\begin{table}[H]51\centering\footnotesize52\caption{Construction of the universe and of the primary sample}53\label{tab:universe}54\begin{minipage}[t]{0.42\textwidth}\centering55\input{tables/tab02_universe}56\end{minipage}\hfill57\begin{minipage}[t]{0.56\textwidth}\centering58\label{tab:cascade}59\input{tables/tab02b_sample_cascade}60\end{minipage}61\begin{minipage}{\textwidth}\vspace{0.3em}\footnotesize62\emph{Notes.} Left: firms. Right: event-time observations (five horizons pooled); steps are applied in the frozen order63of \texttt{docs/PIT\_PROTOCOL.md}.64\end{minipage}65\end{table}6667\begin{table}[H]68\centering\footnotesize69\caption{Summary statistics, primary cohort ($h=20$)}70\label{tab:summary}71\adjustbox{max width=\textwidth}{\input{tables/tab03_summary}}72\begin{minipage}{\textwidth}\vspace{0.3em}\footnotesize73\emph{Notes.} Intraday families are daily values averaged over the 20-session window ending at the cutoff; realized74variances are sums of squared bar log-returns within the regular session. Raw, unwinsorised values.75\end{minipage}76\end{table}7778\begin{figure}[H]79\centering80\includegraphics[width=\textwidth]{figures/fig02_coverage.pdf}81\caption{Sample coverage and adverse-outcome rate by year of cutoff ($h=20$ cohort).}82\label{fig:coverage}83\end{figure}84