Weighted-Average Portfolio Expected Return

Why Portfolio Expected Return Is a Weighted Average

An investor’s portfolio combines assets in different amounts. Its expected return therefore depends both on the expected return of each asset and on how much of the portfolio is allocated to that asset. An asset with a high expected return has little influence if it represents only a small share of the portfolio; an asset held in a larger proportion has more influence.

The key calculation is a weighted average, but its use depends on defining the weights and the return period consistently. This tutorial derives the formula for a one-period portfolio, applies it to scenario-based returns, and shows how shifting an allocation affects expected return. The calculation concerns the portfolio’s expected return, not its variability or realized return over a sequence of changing allocations.

Definition: For asset $j$, the portfolio weight $w_j$ is the amount invested in that asset at the beginning of the period divided by the portfolio’s total beginning value. The expected portfolio return is the probability-weighted average of the portfolio return across possible outcomes for the period.

The Model and the Weighted-Average Formula

Suppose a portfolio holds $n$ assets over the same single period. Let $V_0$ be its total value at the beginning of the period, let $w_j$ be the beginning-value weight of asset $j$, and let $R_j$ be that asset’s simple return during the period. Assume there are no external deposits or withdrawals during the period, and that the portfolio holdings are not changed during it. Ignore fees and taxes unless returns are explicitly stated net of those costs. The weights satisfy $\sum_{j=1}^n w_j=1$ for a fully invested portfolio; an allocation to cash can be included as another asset.

Under these assumptions, the portfolio’s one-period return $R_P$ is the sum of the asset returns multiplied by their beginning weights. Taking expectations gives the expected portfolio return.

R_P=\sum_{j=1}^{n}w_jR_j \qquad\text{and}\qquad E[R_P]=\sum_{j=1}^{n}w_jE[R_j].

Here $E[R_j]$ is the expected return of asset $j$ for the same period. The expected returns can be obtained from scenarios or supplied directly, but their horizon and return definition must match the portfolio calculation. For example, one-period simple returns should not be mixed with annualized returns for a different horizon.

Why the Formula Holds

The result follows from how portfolio value changes. It does not require the assets to have independent returns.

Write each initial investment.
Asset $j$ begins with $w_jV_0$ dollars invested. After its one-period return $R_j$, that holding is worth $w_jV_0(1+R_j)$.

Add the ending values.
The ending portfolio value is $\sum_{j=1}^{n}w_jV_0(1+R_j)$. Since the weights sum to one, this equals $V_0\left(1+\sum_{j=1}^{n}w_jR_j\right)$.

Calculate the portfolio return.
Subtract the initial value $V_0$ from the ending value and divide by $V_0$. The result is $R_P=\sum_{j=1}^{n}w_jR_j$.

Take expectations.
Expectation is linear: $E[\sum_j w_jR_j]=\sum_j w_jE[R_j]$ when the weights are fixed. This identity does not require independence among asset returns.

The derivation relies on fixed beginning weights over the stated period and a portfolio return formed from the holdings’ values. If the portfolio is rebalanced during the period, or cash flows occur, the simple one-period calculation needs to account for those details. Also, weights need not all be positive: a short position can have a negative weight. In a leveraged portfolio the weights can include values above one, with negative weights offsetting them; the formula still applies if the positions and portfolio return are defined consistently.

Worked Example: Combining Three Scenario-Based Expected Returns

A hypothetical investor places $40,000 in asset A, $35,000 in asset B, and $25,000 in a cash account. The period is one year, with no deposits, withdrawals, fees, or taxes. Asset A returns 12% or −4%, each with probability 0.50. Asset B returns 8% with probability 0.25 and 2% with probability 0.75. The cash account earns a certain 2% for the year. Find the portfolio’s expected return and expected dollar gain. Use the stated discrete distributions for A and B; no assumption about dependence between them is needed.

Step 1: Find the beginning weights. The total initial value is $40{,}000+35{,}000+25{,}000=\$100{,}000$. Therefore, $w_A=40{,}000/100{,}000=0.40$, $w_B=35{,}000/100{,}000=0.35$, and $w_C=25{,}000/100{,}000=0.25$. They sum to $0.40+0.35+0.25=1.00$.

Step 2: Calculate each expected return. For A, $E[R_A]=0.50(12\%)+0.50(-4\%)=6\%-2\%=4\%$. For B, $E[R_B]=0.25(8\%)+0.75(2\%)=2\%+1.5\%=3.5\%$. The cash account is certain to return $E[R_C]=2\%$.

Step 3: Weight the expectations. The portfolio’s expected return is $0.40(4\%)+0.35(3.5\%)+0.25(2\%)=1.60\%+1.225\%+0.50\%=3.325\%$.

Step 4: Convert to an expected dollar gain and check. The expected gain is $0.03325(\$100{,}000)=\$3{,}325$. Dividing that gain by the initial portfolio value gives $\$3{,}325/\$100{,}000=0.03325=3.325\%$, which agrees with the weighted-return calculation.

Method and mistake avoided: First find each asset’s expected return, then apply the beginning portfolio weights. An unweighted average of the three expected returns would give $(4\%+3.5\%+2\%)/3=3.167\%$ (rounded), which is wrong because the investor did not allocate equal amounts to the three holdings.

How Allocation Changes Expected Return

For two assets with expected returns $\mu_A=E[R_A]$ and $\mu_B=E[R_B]$, let $x$ be the fraction invested in A and $1-x$ the fraction invested in B. Provided the weights remain fixed over the period, expected portfolio return is $x\mu_A+(1-x)\mu_B$. If the allocation to A changes by $\Delta x$, with the same amount transferred out of B, then the change in expected return is $\Delta x(\mu_A-\mu_B)$.

E[R_P]=x\mu_A+(1-x)\mu_B $$ $$ \Delta E[R_P]=\Delta x(\mu_A-\mu_B).

This comparison isolates the effect of reallocating between the two assets: increasing the share of the asset with the higher expected return raises expected portfolio return, while increasing the share of the lower-return asset reduces it. The formula does not say that the higher-return asset is preferable overall; it describes the effect on expected return only.

Worked Example: The Effect of Moving Capital Between Two Assets

A hypothetical $100,000 portfolio holds asset A and asset B for one year, with no interim trading or cash flows. Asset A returns 18% or 0% with probability 0.50 each. Asset B returns 2% with probability 0.25 and 6% with probability 0.75. The investor considers raising A’s allocation from 60% to 75%, transferring the difference from B. Calculate the expected return before and after the change and the change in expected dollar gain. Treat the scenario probabilities as the return distributions for this one-year period.

Step 1: Find the assets’ expected returns. For A, $E[R_A]=0.50(18\%)+0.50(0\%)=9\%$. For B, $E[R_B]=0.25(2\%)+0.75(6\%)=0.5\%+4.5\%=5\%$.

Step 2: Calculate the original portfolio expectation. Its weights are 0.60 in A and 0.40 in B, so $E[R_{P,\text{old}}]=0.60(9\%)+0.40(5\%)=5.4\%+2.0\%=7.4\%$.

Step 3: Calculate the new portfolio expectation. Its weights are 0.75 in A and 0.25 in B, so $E[R_{P,\text{new}}]=0.75(9\%)+0.25(5\%)=6.75\%+1.25\%=8.0\%$.

Step 4: Check the allocation effect. The shift is $\Delta x=0.75-0.60=0.15$, and the expected-return difference between the assets is $9\%-5\%=4$ percentage points. Thus $\Delta E[R_P]=0.15(4\%)=0.60$ percentage points, matching $8.0\%-7.4\%=0.6$ percentage points. The expected dollar gain rises from $0.074(\$100{,}000)=\$7{,}400$ to $0.08(\$100{,}000)=\$8{,}000$, a difference of $600.

Method and mistake avoided: Apply the transfer to both weights: an increase of 15 percentage points in A means a decrease of 15 percentage points in B. Keeping B at 40% while raising A to 75% would produce weights summing to 115%, not the stated fully invested portfolio.

Worked Example: Calculating Portfolio Expectation by Scenarios

A hypothetical investment committee considers three assets over one year. The economic scenario is strong with probability 0.20, steady with probability 0.50, and weak with probability 0.30. Returns are as follows: asset A returns 20%, 8%, and −10% in those scenarios; asset B returns 5%, 5%, and 5%; asset C returns −5%, 10%, and 15%. The portfolio weights are 0.50, 0.30, and 0.20, respectively. Assume no trading or cash flows during the year. Calculate the portfolio expected return by first finding each asset expectation, then verify it directly across scenarios.

Step 1: Find the expected return of each asset. For A, $E[R_A]=0.20(20\%)+0.50(8\%)+0.30(-10\%)=4\%+4\%-3\%=5\%$. For B, $E[R_B]=0.20(5\%)+0.50(5\%)+0.30(5\%)=5\%$. For C, $E[R_C]=0.20(-5\%)+0.50(10\%)+0.30(15\%)=-1\%+5\%+4.5\%=8.5\%$.

Step 2: Weight the asset expectations. The expected portfolio return is $0.50(5\%)+0.30(5\%)+0.20(8.5\%)=2.5\%+1.5\%+1.7\%=5.7\%$.

Step 3: Check by calculating the portfolio return in each scenario. In the strong scenario it is $0.50(20\%)+0.30(5\%)+0.20(-5\%)=10\%+1.5\%-1\%=10.5\%$. In the steady scenario it is $0.50(8\%)+0.30(5\%)+0.20(10\%)=4\%+1.5\%+2\%=7.5\%$. In the weak scenario it is $0.50(-10\%)+0.30(5\%)+0.20(15\%)=-5\%+1.5\%+3\%=-0.5\%$.

Step 4: Take the probability-weighted average of those portfolio returns. The result is $0.20(10.5\%)+0.50(7.5\%)+0.30(-0.5\%)=2.1\%+3.75\%-0.15\%=5.7\%$, agreeing with Step 2. The scenario describes the assets jointly, but the equality does not depend on their being independent.

Method and mistake avoided: Use the same scenario probabilities and portfolio weights throughout. A common error is to average scenario returns equally; here the scenario probabilities are 0.20, 0.50, and 0.30, not equal.

Common Traps and Interpretation

The weighted-average formula is compact, but exam calculations can go wrong if weights, returns, or periods are inconsistent. A useful check is to confirm that the weights represent the portfolio at the beginning of the relevant period and sum to one for a fully invested portfolio. Then confirm that every asset return covers the same period and uses the same definition, such as simple return.

Using equal weights without justification. The arithmetic mean of asset expected returns is appropriate only when the asset weights are equal. Otherwise, multiply each expectation by its actual portfolio weight.
Failing to adjust the other holdings. If a fixed amount is transferred into one asset in a fully invested portfolio, the total allocation elsewhere must fall by the same amount.
Mixing units. Keep returns consistently as percentages or decimals. For example, $0.40(4\%)+0.35(3.5\%)$ is valid; multiplying 0.40 by 4 and interpreting the result as a decimal return is not.
Assuming independence is required. The expected-return formula uses linearity of expectation. Dependence among asset returns matters for measures of portfolio risk, but not for this weighted expected-return calculation.
Applying beginning weights to a different setup. Rebalancing during the period, contributions, withdrawals, fees, or taxes can change the return calculation. State the applicable period and assumptions rather than treating every portfolio as a static one-period portfolio.

Expected return is a summary of possible returns, not a promise that the portfolio will earn that amount. The weighted-average calculation is useful for comparing allocations and translating asset-level expectations into a portfolio-level expectation. It should be interpreted alongside other considerations, including risk, costs, and the suitability of the assumptions used to estimate each asset’s expected return.

Key takeaway: For fixed portfolio weights over a common period, expected portfolio return equals the sum of each beginning weight multiplied by that asset’s expected return. Changing an allocation changes expected return according to the weight shifted and the difference between the expected returns of the assets involved.

Check Your Understanding

For each problem, use the stated one-period assumptions and check that the weights and return horizons are consistent.

A fully invested portfolio assigns 70% to an asset with expected return 6% and 30% to an asset with expected return 2%. Find the portfolio expected return. (Answer: 4.8%)
A portfolio holds three assets with weights 0.25, 0.45, and 0.30 and expected returns 10%, 4%, and 1%, respectively. Find its expected return and verify the weights sum to one. (Answer: Weights sum to 1.00; expected return is 4.6%)
A $60,000 portfolio has 40% in an asset with expected return 7% and 60% in an asset with expected return 3%. The investor transfers 10% of the portfolio value from the second asset to the first. Find the new expected return and the change in expected dollar gain. (Answer: New expected return is 4.6%, an increase of 0.4 percentage points; expected dollar gain rises by $240.)
A portfolio invests 50% in A and 50% in B. A returns 12% or 0% with equal probabilities; B returns 8% with probability 0.25 and 4% with probability 0.75. Find each asset’s expected return and the portfolio expected return. (Answer: $E[R_A]=6\%$, $E[R_B]=5\%$, and $E[R_P]=5.5\%$.)
A fully invested portfolio allocates 40% to B, with expected return 5%. The remaining amount is split between A, with expected return 10%, and C, with expected return 2%. What weight in A is required for a portfolio expected return of 6.4%, and what is the resulting weight in C? (Answer: A weight is 40%; C weight is 20%.)

What You'll Learn