Skip to content
All library documents

In-Sample and Out-of-Sample Testing for Mean-Variance Portfolios

Article Quant Q&A · Author: renato
Read the original

Summary

The document describes a proposed evaluation of mean-variance portfolios using daily S&P 500 returns across a three-year dataset. The researcher estimates optimal weights, including global minimum-variance and tangency portfolios, from the first two years, then considers measuring portfolio returns and risk over a later one-year period. They call this an in-sample test, although the evaluation period follows the estimation window.

For out-of-sample evaluation, the proposal starts with the later year and rolls the estimation window forward, rebalancing daily without transaction costs. The post contains the researcher's setup and uncertainty, but no answer or performance results. It therefore serves as a methodological question rather than a validated testing prescription; the limited sample and omitted trading costs are explicit constraints.

Key ideas

  • The proposed optimization estimates portfolio weights from an initial two-year return sample.
  • The subsequent year is used to observe portfolio performance and risk.
  • The proposed rolling evaluation recalculates weights and rebalances daily over the later period.
  • The researcher acknowledges the short dataset and assumes no transaction costs.

Tags

Portfolio constructionBacktestingUS markets
Full text
# In sample and out of sample in Mean Variance Optimization


# In sample and out of sample in Mean Variance Optimization












Hello to everyone and thanks again for your help, i have find this forum really helpful while working on my final dissertation.

However I'm here again because I have loads of doubts regarding the in-sample and out of sample test of my model. I will explain from my point of view and I would like to know from you if I am completely wrong or if there is something true in what i believe. So my setup is:

- my dataset is made of daily return of the S&P500 from 1st of January 2016 till the 31st December 2018

- I have performed my optimization on the first two years 1st January2016 till 31st December 2017 and found the various optimal weights i.e Global mean variance portfolio, tangency portfolio, and few other constrained portfolios.

- Now suppose i want to perform an insample test and an out of sample test, both of 1 year period, on this hand i wanted to proceed as follows:

- for the in-sample test i use the optimal weights i have found with my sample of 2 years, and I see how these portfolios perform using the return of the year 2017-2018. then I find the expected returns, and the measure of risk from these portfolios observation.

- for the out of sample instead, the method is to use 1 year of observation as starting point the year 2017-2018 and moving with a rolling window rebalancing my portfolio every day for 1 year (i assume no transaction cost)

Now I am wondering if this could make any sense, every help and suggestion will be accepted. Ps i know that i don't have enough data to perform the optimization, but for the moment i can't change my dataset.

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.