Volume-Weighted Duration as an Intraday Activity Measure
Summary
The document considers how to quantify intraday market activity from tick-by-tick order book events, especially when trading conditions change around events such as earnings announcements. It points to volume-weighted duration: measure how long it takes the market to exchange a fixed quantity of shares. Shorter durations indicate that the quantity is being traded more quickly, providing one way to track changes in activity.
The answer also directs readers to autoregressive conditional duration models, which model the timing between market events and are used in research on intraday trading activity. It cites prior literature and a survey as starting points, but presents no implementation details, comparison with alternatives, or empirical result for a particular instrument. The proposed measure focuses on trading volume and elapsed time; it does not by itself summarize every type of order book change, such as additions and cancellations.
Key ideas
- Volume-weighted duration measures the time required to trade a fixed quantity of shares.
- Faster completion of that quantity can indicate greater intraday trading activity.
- Autoregressive conditional duration models provide a framework for modeling event timing.
- The suggested duration measure does not capture all order book changes directly.
Tags
Full text
# Metrics for Measuring Intraday Activity # Metrics for Measuring Intraday Activity Using intraday data, are there certain measures that exist for measuring the "activity" of a particular instrument. By intraday data I mean, all changes to the limit order book (trades, adds, cancels, etc.), tick by tick. For example, let's say on any average day, this "measure of activity" for Apple stock would be relatively similar, but on a day where Apple is making an earnings release or some big event, I'd expect there to be more trading activity (maybe more adds/cancels, etc.) . Are there any particular measures that can help me determine this? ## Answer by Malick (score 1) https://quant.stackexchange.com/a/36410 There is an interesting literature about volume weighted duration. The idea is to measure the activity of a market by measuring the time necessary to exchange a fix amount of stocks. This can be seen as a measure of intra-day activity. You can have a look to these papers : - Gouriéroux, C., Jasiak, J., & Le Fol, G. (1999). Intra-day market activity. Journal of Financial Markets, 2(3), 193–226. http://doi.org/10.1016/S1386-4181(99)00004-X item - Hautsch, N. (2001). Modelling Intraday Trading Activity Using Box-Cox ACD Models. working paper. Retrieved from http://www.ssrn.com/abstract=289643 These models are generally based on Autoregressive Conditional durations models (ACD), see the following excellent survey to get an overview : - Pacurar, M. (2008). Autoregressive conditional Duration (ACD) models in finance: a survey of the theoretical and empirical literature. Journal of Economic Surveys, 1(February). Retrieved from http://onlinelibrary.wiley.com/doi/10.1111/j.1467-6419.2007.00547.x/full
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.