quantverse.
data / bars / px.eq.us.bars.dollar

Dollar bars px.eq.us.bars.dollar

Volume-weighted bars sampled at fixed dollar thresholds.

PIT · Pro tier · 2008-01 → today · 1.2B rows

Coming soon — not served by the API yet.

Methodology

A bar closes when cumulative traded notional crosses the threshold, not when the clock ticks over, so each bar carries roughly the same amount of market activity and the sampling frequency follows volatility instead of imposing it. Thresholds are set per symbol from its own trailing notional so a large cap and a small cap produce comparable bar counts. start_ts and end_ts are both kept: a bar that took nine minutes and one that took nine seconds are different evidence, and a strategy that reads only the close of the bar is reading information from an unknown point in time.

Schema

ColumnTypeRole
as_of_tstimestampknowledge time · UTC
effective_tstimestampbar close time · US/Eastern
symbolstringticker in force on effective_ts
bar_indexint64ordinal within the symbol
start_tstimestampfirst print in the bar
end_tstimestamplast print in the bar
openfloat64first trade price
highfloat64bar high
lowfloat64bar low
closefloat64last trade price
volumeint64shares in the bar
dollar_volumefloat64notional in the bar
trade_countint32prints aggregated

Use cases

  • Train return models on closer-to-normal, less autocorrelated samples.
  • Compare a strategy's statistics against a time-bar baseline.
  • Size bars consistently across symbols with very different liquidity.

Background

Related packages

See pricing · all packages