2-25. Attribution

In this lesson, you learned how to use multi-factor models to decompose portfolio P and L, and risk into components attributable to each factor and offer. This can be useful for understanding and explaining your portfolio’s performance.

We can also decompose portfolio return, also called P&L or profit and loss. If we sum up the asset returns in a weighted sum using the holdings vector as weights, we have the portfolio P&L for one time period. Let’s substitute in the multifactor model expression for r. Here, we can substitute the exposure vector … Read more

When we’ve described our portfolio in terms of several factors, we can also use that information to decompose the portfolio risk in terms of the risk due to these factors, and the idiosyncratic risk due to the error term. Let’s start with the expression for the portfolio variance. First, we write down the variance of … Read more

Let’s say we have a portfolio. That means we’ve calculated a holdings vector. So we have a vector that tells us how much money we want to invest in each asset. The holdings vector is an n-dimensional vector with one value for each company. If we take the dot product of the holdings vector and … Read more

Welcome back. In this lesson, we’re going to learn about attribution, which is a way to understand sources and drivers of risk and return in your portfolio, by breaking down metrics of risk and return into components.