Aligning Volume Bars Across Multiple Futures
Summary
The document asks how to align volume bars from different futures so their returns can be used in multivariate analysis. It notes that each contract reaches a fixed traded-volume threshold at different times, making timestamps inconsistent, and cites volume bars as a way to represent trading activity more directly than fixed time bars. The discussion presents two possible approaches rather than a settled standard.
One answer proposes setting each bar to a fixed share of that asset’s daily volume, so each asset produces the same number of bars per day. Another suggests generating bars jointly, closing a bar when the combined volume across the assets reaches a threshold. The document offers no comparison, empirical results, or detailed guidance on choosing thresholds. The methods therefore describe candidate alignment conventions, with different implications for how each bar represents market activity; researchers would need to assess which convention suits their analysis.
Key ideas
- Volume bars for separate futures usually close at different times because their trading volumes differ.
- A fixed fraction of each asset’s daily volume can produce the same number of bars per day.
- A joint threshold based on combined asset volume is another proposed way to align bars.
- The discussion does not establish which alignment method performs better.
Tags
Full text
# Align volume bars for multivariate analysis # Align volume bars for multivariate analysis Looking at the book "Advances in financial machine learning" the author proposes a way to sample high frequency financial data in several fashions which are not only the standard time bars. I was trying to prepare a dataset for doing multivariate analysis using time series of several different futures. In doing that I would use volume bars, which are considered more representative of the effective flow of information provided by the market and also have desirable statistical properties like the recovering of normality, just looking at the returns. Unfortunately, when I create volume bars for different assets given a threshold of traded contracts that signals to close the bar, the resulting bars are obviously not aligned. Actually, every future contract has its own volume path during the day, so there is no chance to get them aligned in time. My concern comes when one wants to do some multivariate regression between the assets considered. How do I solve the problem of differently aligned timestamps? Is there a common way to solve this in practice? I suspect that working with volume bars (or even tick/dollar bars) implies that you have to deal with the timestamp issue. ## Answer by chrisaycock (score 1) https://quant.stackexchange.com/a/54550 It's not about timestamps. You just need to assign the same meaning to each bar. Choose a fixed percentage of daily volume each bar should represent. Then for each individual day, compute the bar size from that percentage: ``` today's bar size = today's total volume * chosen fixed percentage ``` For example, I can choose that each bar should be 2% of the daily volume. So for each day, I will have a different bar size to get that 2%. And voila, I'll have the same number of bars within each day. ## Answer by ZhangYou0122 (score 0) https://quant.stackexchange.com/a/75730 I'm also wondering about the same issue. There is even an exercise(2.3) that requires readers to "Apply ETF tricks" "On DOLLAR BAR series of E-mini S&P 500 futures and Eurostoxx 50 futures". There're are few code implementations floating around on Github, however, all of them assume that the bars are already aligned. I think the alignment should be done when generating the dollar bars, the volumes of all assets need to be considered in the process. For example, we could generate a dollar bar for each of the assets whenever the sum of asset volumes reaches a certain threshold.
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.