Skip to content
All library documents

Choosing Price Data for Mean-Variance Portfolio Analysis

Article Quant Q&A · Author: dynra

Summary

The question asks whether Markowitz portfolio construction should use closing prices or dividend-adjusted prices when estimating a minimum-variance portfolio, an optimal risky portfolio, and an efficient frontier. The answer frames the choice around how the portfolio will be monitored: closing prices may suit end-of-day or snapshot-based tracking, while adjusted prices are suggested for intraday price data. It also notes that dividends can be incorporated through adjusted stock prices.

The response offers only brief guidance and does not lay out a complete return-calculation procedure or explain the adjustment conventions. It mentions that option underlyings may require a different treatment, such as risk-neutral pricing depending on the strategy used. It does not provide data tests, portfolio estimates, or a worked frontier example. Therefore, the main takeaway is to align the price series with the portfolio’s tracking frequency and dividend treatment, while recognizing that the answer leaves important implementation details unresolved.

Key ideas

  • Price-series choice should match how frequently the portfolio is tracked.
  • Closing prices may be appropriate for end-of-day or snapshot-based portfolio monitoring.
  • Adjusted prices can account for stock dividends in return analysis.
  • Option underlyings may need valuation treatment suited to the instrument and strategy.
  • The brief answer does not specify a full return-estimation or dividend-adjustment workflow.

Tags

Full text
# Answer by Ragu (score 0)


# What price data should I used when making minimum mean variance portfolio, optimal risky portfolio and efficient frontier using Markowitz?












I need to make optimal risky portfolio, minimum variance portfolio and efficient frontier using Markowitz . But i don't know whether to used close price data or adjusted data. If i'm using adjusted data, did i have include some additional formula for the dividend?

## Answer by Ragu (score 0)

https://quant.stackexchange.com/a/37229

It depends on how your portfolio needs to be tracked, if it's end-of-day or snapshot driven portfolio, you could take the close price data from any of the exchange traded. Else, adjusted data intraday prices.

For minimum variance portfolio, i suppose you take the vertex [x0 + a, y0] coordinates from hyperbola of set of possible mean-variance patterns from your set of possible underlyings. Dividends could be adjusted with the stock price, if underlying is call option, you may need to adjust with risk neutral pricing or what ever strategy you are using to derive the underlying prices.

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.