Re: Average Inflation Question

A stil better way is to compute the geometric mean of the product of
the Paasche and Laspeyres indices, which yields the Fisher Ideal Index.

http://en.wikipedia.org/wiki/Price_index

Yet even better than the Fisher Ideal is the Divisia index, but wiki
doesn't have the cite for it. An advantage of the Divisia over the
Fisher is that it is easier to identify specific components that have a
large effect on the index.

If memory serves, the book "Index Numbers in Theory and Practice" by
RGD Allen covers this material.

Regards,

Bruce

B D McCullough
Professor of Decision Sciences
Drexel University


espyrian wrote:
1_Patriotic_Guy wrote:
If I know inflation between year one and year two, and between year two
and year three, and so on -- How do I calculate the average inflation
over a multi-year period -- There has to be a formula which involves
plugging in the annual inflation rate for each and every
year. Below is my best guess at an answer, but how do I
handle input negative inflation years (e.g. just inputting a negative
number doesn't seem like the right approach, doesn't seem to properly
influence the average.)
Convert each inflation rate to a factor, so 3.5% becomes 1.035,
2.3% becomes 1.023, and so on. Multiply all N factors together and

then take the Nth root of the product. Then convert the factor so

obtained back to a percentage in the obvious way.

That is an excellent best guess. Yes, the correct approach is to use
the geometric mean (http://en.wikipedia.org/wiki/Geometric_mean).

"The geometric mean is useful to determine 'average factors'. For
example, if a stock rose 10% in the first year, 20% in the second year
and fell 15% in the third year, then we compute the geometric mean of
the factors 1.10, 1.20 and 0.85 as (1.10 × 1.20 × 0.85)^1/3 =
1.0391-1... and we conclude that the stock rose 3.91 percent per year,
on average." -Wikipedia

.