Index-number formula

There are a great many index-number formulas that can be used to combine information on prices at two points in time to produce a price index, despite the common goal of measuring price changes over time. This section of the course covers 10 important index-number formulas that are used in practice to calculate a price index. Associated with each formula is the name of the person who developed it. It is important to associate the names with the formulas, as they are usually referred to by name in application. This is unfortunately an exercise is memorization, as most of the index-number formulas have a similar form.

To help build intuition and a deeper understanding of what the index-number formulas are doing, these different approaches for calculating a price index can be broadly classified as either arithmetic price indices or geometric price indices. Not all index-number formulas fall into these two groups—for example, there are also harmonic price indices, a category which can overlap with arithmetic indices—but arithmetic and geometric indices are fairly easy to understand, with most index-number formulas falling into one of these two categories.

In practice a price index is multiplied by 100, so that the percent change in the index value over two periods is simply the index value minus 100. This is just a convenient normalization that has no bearing on the economic content of a price index, and is ignored in this section. Note that all the index-number formulas can be multiplied by 100 to put them on the usual scale.

📖 PPI manual: Chapter 1, sections B1–B3.