Why Year-to-Date Averages Are Not Year-over-Year Comparisons

Summary

The document distinguishes a full-year monthly average from an average based on only the first few months of the following year. Comparing these figures as a percentage change does not make the result a year-over-year comparison, because the periods differ in both length and seasonal coverage. It also warns that averaging volatile observations across periods of unequal length can be misleading.

One possible estimate is to assume the latest observed monthly figure persists through the rest of the year. The answer identifies this as a carry-over approach, used in some economic calculations, but emphasizes that the assumption may become inaccurate as later observations arrive. The example is about sales volume, not a trading indicator or market strategy. It offers no statistical test or evidence that the carry-over assumption will hold; the comparison should be described according to the periods actually measured, and treated cautiously as a projection rather than an observed annual result.

Key ideas

• Comparing a full-year average with a partial-year average is not a year-over-year comparison.
• Unequal observation periods can make averages difficult to interpret.
• A carry-over estimate assumes the latest observed level persists through the remaining period.
• The carry-over assumption is uncertain because later values may change.

Tags

# YOY Comparison?


# YOY Comparison?












I have 2 numbers:

Average monthly sales volume for 2015 (Jan to Dec): 1,000

Average monthly sales volume for 2016 YTD (Jan and Feb): 1,200

If I want to compare these 2 numbers with a % change, is it a YOY comparison?

Or, is it a 2015 year-end average compared to 2016 year-to-date average?

## Answer by Hector (score 0, accepted)

https://quant.stackexchange.com/a/24437

You can compute a variation by making assumptions. The simplest way is to assume that the last known figure (here, your monthly sales volume for jan-feb 2016) is going to stay constant until the end of the period. This calculation is often used to compute GDP variations and is called carry-over. However, you should consider that this figure is not really significant: a lot can happen until the end of the year.

The result will not be a Year on Year calculation, and I do not believe that a year average vs 2 months average makes much sense as you compared potentially volatile figures observed over periods with different lengths.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.