Estimating Stock Cost-Basis Distributions with a Turnover-Decay Model

Summary

This article describes a model for estimating how a stock’s traded shares are distributed across historical purchase prices, a concept often called chip distribution in Chinese market analysis. Since investor-level holdings and cost bases are not public, the method uses historical volume, turnover, and price data to approximate the distribution. It assumes shares at different price levels have equal chances of trading on a later day, with that day’s turnover serving as the fraction that changes hands. Earlier trades are therefore progressively discounted by cumulative turnover-based retention factors.

The process truncates the history after cumulative turnover approaches a near-complete trading cycle, uses daily closes as representative trade prices, and bins prices to build a distribution. It derives measures such as average cost, profitable-share estimates, concentration ranges, kurtosis, and the most concentrated price bin. These are descriptive technical features, not observed account holdings. The article offers no predictive validation, and its equal-turnover and daily-close assumptions can misstate actual ownership and transaction prices; it also does not model company fundamentals.

Key ideas

• The model estimates cost-basis distributions from public price, volume, and turnover data because account-level holdings are unavailable.
• It treats daily turnover as the fraction of shares that changes hands and discounts older trades by cumulative retention.
• The method limits its history to a near-complete turnover cycle and represents each day’s trades with the close.
• Price bins support concentration, average-cost, profitable-share, and kurtosis measures.
• The estimates depend on simplifying assumptions and do not establish predictive performance or incorporate fundamentals.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.