Portfolio Lab

Portfolio analysis step by step: what you own and what drives it

Basics9 min read · updated 19 November 2026 · 8 sources

A portfolio analysis answers two questions: what do you really own, and what drives the result? The list at your broker only shows positions and prices. It does not show that the same stock sits in three funds, that a small position carries most of the volatility, or what a year like 2008 would do to your wealth. This article walks through an analysis in six steps and works them out for a sample portfolio of $100,000. Not investment advice.

What a portfolio analysis is for

A portfolio rarely grows according to plan. It starts with a world ETF; later a thematic fund, a few individual stocks and perhaps a second ETF from a new broker are added. Each decision made sense on its own. After a few years, often nobody knows the whole exactly. Research shows that this is no exception: according to Goetzmann and Kumar (2008), many individual investors diversify too little, and French and Poterba (1991) describe how strongly investors overweight their own market (see home bias).

A portfolio analysis needs no forecasts. It measures what is there and turns it into six answers:

  1. Holdings: Which positions are there, and how much does each one weigh?
  2. Costs: What does the portfolio cost per year in ongoing fees?
  3. Look-through: Which countries, sectors and individual stocks are inside the funds?
  4. Risk: Which position contributes how much to the volatility?
  5. Stress test: What happens in a bad year?
  6. Factors: Which characteristics of the stocks explain the result?

The order is deliberate. The first three steps need only the weights and the composition of the funds, so no price history. Volatility and co-movement only come in for risk, stress test and factors.

Steps 1 and 2: holdings and costs

It starts with a table of today's value of each position. What counts is the market value, not the purchase price. Accounts at several brokers belong in the same table, as do funds from a savings plan that are easy to forget. The example for this article:

PositionValueWeightOngoing costs
World ETF (developed market stocks)$50,00050%0.20%
Technology ETF (US)$20,00020%0.30%
Individual stock (large US technology company)$15,00015%none
Government bond ETF$15,00015%0.15%

The costs of the portfolio are the rate of the positions averaged with their weights. Individual stocks have no ongoing costs, but buying and selling them costs fees. A first measure of diversification is the effective number: it says how many equally sized positions the portfolio corresponds to.

weight of position i as a share of the portfolio (0.5 for 50%)

Step 3: look inside the funds

An ETF is a wrapper. If you want to know what you own, you have to open the wrapper and add up the positions of all funds. This approach is called look-through. The same calculation applies to every country, every sector and every individual stock:

share of the fund in the portfolio
weight of the country, sector or stock in the fund

Rounded assumptions apply to the example: the world ETF holds 70% of its stocks in the United States, the technology ETF 100%. The individual stock is also the largest position of both funds, at 5% in the world ETF and 9% in the technology ETF.

Two more measures belong to this step. The overlap of two funds is the sum of the smaller weights of all stocks they share; a value of 40% means that 40% of both funds are identical. And the share of the ten largest stocks shows how much the portfolio depends on a few companies. Thematic and sector funds often overlap strongly with the world ETF you already hold (see thematic ETFs).

Step 4: who carries the risk?

Weights say where the money is. They do not say where the volatility comes from. A position with high volatility that moves in step with the rest carries more risk than its weight suggests. The decomposition following Litterman (1996) assigns each position its share of the portfolio's volatility, and the shares add up to 100% (Maillard et al. (2010); see risk contribution).

weight of position i
covariance matrix of the returns
covariance of position i with the whole portfolio
variance of the portfolio

Assumptions for the example: annual volatility of 15% for the world ETF, 22% for the technology ETF, 35% for the individual stock and 5% for the bonds. Correlations: 0.9 between the two equity ETFs, 0.7 between the world ETF and the individual stock, 0.8 between the technology ETF and the individual stock, 0.1 between the bonds and everything else.

Weight and share of risk by position

Worked example of the article with assumed volatilities (15, 22, 35 and 5%) and correlations, not market data. The share of risk follows the formula above.

The numbers depend on the assumptions, the pattern does not: volatile positions that move with the rest carry more risk than weight, calm and weakly correlated ones less (see correlation and diversification). Risk also includes currency. The stocks outside the United States in the world ETF are held in other currencies; what that means is explained in the article on currency risk.

Step 5: stress test

Volatility describes ordinary years. A stress test asks about the unusual ones: what does the portfolio lose if a crisis repeats itself or an assumed scenario occurs? The simplest form gives each position a shock and adds up. Suppose the world ETF loses 40%, the technology ETF 50% and the individual stock 60%, and the government bonds gain 5%.

It becomes more precise with historical crises and with shocks to risk factors such as interest rates, the stock market and the oil price, where the other variables move along according to their usual relationship (see stress test with historical crises and scenario analysis). The measure for the worst decline of the past is the maximum drawdown.

Step 6: factors and style

The last step asks which characteristics of the stocks explain the result. According to Fama and French (1993), the returns of stock portfolios can largely be traced back to a few factors: the market, the size of the companies and their valuation (see Fama-French factors). Sharpe (1992) shows how to infer a fund's investment style from its returns (see style analysis).

For the sample portfolio the result is predictable: market risk above that of the world market, large companies and a clear tilt toward expensively valued growth stocks. As long as this style is in favor, the portfolio looks better than the market. When the wind turns, the same tilt explains the shortfall. Factor analysis protects against mistaking a style bet for skill in selection.

What follows from the analysis

Often nothing follows, and that is a good result: the portfolio is as diversified as intended. If the analysis shows concentrations, there are three ways to deal with them.

  • Keep it deliberately: A concentration is not a mistake if you know it and can bear the loss in the stress test. According to Bessembinder (2018), the wealth created by the stock market comes from a few companies; betting on individual stocks can hit them or miss them.
  • Steer with new money: Contributions and distributions flow into the underweighted parts. That triggers no taxes and works surprisingly fast while you are still saving regularly.
  • Reallocate: Selling at a gain in a taxable account triggers capital gains tax. A sale is therefore most worthwhile where the risk falls the most, in the example the individual stock (see rebalancing).

A simple rule of your own helps, for example an upper limit for a single stock after look-through. Statman (1987) studied how many stocks are needed for good diversification: clearly more than the ten often mentioned. For most portfolios a broad core takes care of this, with room for tilts around it (see core-satellite).

Limits and common mistakes

  • Looking at the portfolio alone: Your wealth also includes savings accounts, your home, workplace retirement plans and your own earning power. If you work in the industry your portfolio is concentrated in, you carry the risk twice.
  • Counting positions instead of looking inside: Ten funds can diversify less than one if they hold the same stocks.
  • Purchase prices instead of market values: The analysis works with what the positions are worth today.
  • Measures from a short period: Volatility and correlation from two calm years understate the risk. Long periods and a stress test belong in the picture.
  • Forgetting the date: The composition of funds changes. A look-through is a snapshot with a date.
  • Turning every deviation into a trade: Not every number calls for action. Frequent reallocation costs fees and taxes.

How Portfolio Lab calculates

The look-through replaces every fund in your portfolio with its holdings and shows regions, countries, sectors, the largest individual stocks across all funds, the share of the ten largest, the effective number and the overlap of the funds. The stress test calculates the risk shares per position, value at risk, and historical and custom scenarios. The factors tool shows which factor premiums the portfolio carries. All results are model calculations as of the stored fund data; the methodology is linked from each tool.

Frequently asked questions

What is a portfolio analysis?

A review of the whole portfolio: weights, costs, composition of the funds by country, sector and individual stock, the positions' shares of risk and the behavior in crisis scenarios.

How often should I analyze my portfolio?

Once a year is usually enough, plus after larger changes such as a new fund, an inheritance or a strong rise in a single position.

How do I spot concentration risk?

By three numbers after look-through: the share of the largest individual stock, the share of the ten largest stocks and the share of one country or sector. The account overview is not enough, because the same stocks can sit in several funds.

What is the difference between weight and risk share?

The weight says how much money is in a position. The risk share says how much of the portfolio's volatility comes from it. Volatile positions that move with the rest have a higher risk share than weight.

Do a world ETF and a technology ETF overlap?

Yes, usually considerably. The large technology companies are among the largest positions of broad world indexes. Holding both increases their weight in the portfolio.

Do I have to sell after the analysis?

No. It is often enough to direct new contributions into the underweighted parts. Selling at a gain triggers taxes and is most worthwhile where the risk falls the most.

Literature

  1. Bessembinder, Hendrik (2018): Do stocks outperform Treasury bills?. Journal of Financial Economics 129 (3), pp. 440–457. doi:10.1016/j.jfineco.2018.06.004
  2. Fama, Eugene F.; French, Kenneth R. (1993): Common risk factors in the returns on stocks and bonds. Journal of Financial Economics 33 (1), pp. 3–56. doi:10.1016/0304-405x(93)90023-5
  3. French, Kenneth; Poterba, James (1991): Investor Diversification and International Equity Markets. National Bureau of Economic Research, Working Paper 3609. doi:10.3386/w3609
  4. Goetzmann, William N.; Kumar, Alok (2008): Equity Portfolio Diversification. Review of Finance 12 (3), pp. 433–463. doi:10.1093/rof/rfn005
  5. Litterman, Robert (1996): Hot Spots™ and Hedges. The Journal of Portfolio Management 23 (5), pp. 52–75. doi:10.3905/jpm.1996.052
  6. Maillard, Sébastien; Roncalli, Thierry; Teïletche, Jérôme (2010): The Properties of Equally Weighted Risk Contribution Portfolios. The Journal of Portfolio Management 36 (4), pp. 60–70. doi:10.3905/jpm.2010.36.4.060
  7. Sharpe, William F. (1992): Asset allocation. The Journal of Portfolio Management 18 (2), pp. 7–19. doi:10.3905/jpm.1992.409394
  8. Statman, Meir (1987): How Many Stocks Make a Diversified Portfolio?. The Journal of Financial and Quantitative Analysis 22 (3). doi:10.2307/2330969

Full bibliography

This article explains a method; it is not investment or tax advice. Worked examples are simplified and rounded. All terms in the glossary.