posted on 06-06-2019

Stratified Sampling Approach

Updated October 1, 2019

What is a Stratified Sampling Approach?

A stratified sampling approach is an indexing strategy whereby a fund manager divides an index into different "cells" that represent different characteristics of the index. The fund manager then chooses investments that mimic those cells.

How Does a Stratified Sampling Approach Work?

Let's assume you manage the XYZ mutual fund, and you'd like the fund to replicate the S&P 500 index. You could meticulously purchase all the stocks in the index and in the unique quantities that reflect each stock's weight in the index, or you could implement a much simpler stratified sampling approach.

By dividing the securities in the S&P 500 into several categories, like industry, P/E, country, etc. you could then purchase stocks that mimic these characteristics. For example, if 25% of the stocks in the S&P 500 were in the food and beverage industry, you could invest 25% of your fund in food and beverage stocks. These stocks don't necessarily need to be the exact stocks in the S&P 500 index; the objective is simply to mimic the index's characteristics rather than purchase every stock in the index.

Fund managers like the stratified sampling approach because it's easier, faster, and usually involve fewer transactions than fully replicating an index.

But one risk of this approach is that the fund manager could select securities that don't perform as well as those in the index. Another risk is that the manager could choose the wrong characteristics to mimic.

Why Does a Stratified Sampling Approach Matter?

The stratified sampling approach is one of several approaches to managing an index fund.

In general, index performance is hard to replicate because the index doesn't have to deal with capital inflows and outflows like fund managers do. But the goal of the stratified sampling approach is to replicate an index's returns without necessarily replicating the index itself. The securities may differ, but the fund manager is responsible for minimizing the tracking error (that is, the difference between the index's performance and the fund's performance).

It's important to note that the stratified sampling approach generally results in larger holdings of fewer securities than the original index. Thus, the transaction costs and the stock-specific risk are usually higher than the original index.