Portfolio Return¶
One-line definition¶
The weighted return of the assets held in a portfolio for a given period.
Formula¶
r_p = w^T r = sum_i w_i r_i
Units¶
Return, typically decimal or percent.
Why PMs care¶
It is the basic aggregation from security-level performance to portfolio performance.
Worked example¶
Weights 60% / 40%; returns +10% / -5%.
0.6*0.10 + 0.4*(-0.05) = 0.04 = 4%
Code recipe¶
import numpy as np
def portfolio_return(asset_returns, weights):
r = np.asarray(asset_returns, float)
w = np.asarray(weights, float)
return float(w @ r)
Common mistakes¶
- weights not summing to intended net/gross exposure
- mixing percent and decimal return units
- confusing arithmetic period return with multi-period compounding