I am reviewing a work of a colleague of mine.

He graphed two different time series for the same product - the first one (historic figures) takes into account 4 years, while the second one (estimates) has 3 years.

He then added percentages comparisons for the growth rates - almost 15% for the 4-year time series and 16% for the 3-year.

Now I think that comparing 2 time series with unequal lengths might be misleading. For instance, if I use the same length, the new percentages become 10% and 16%, which I think gives a much clearer picture of how the situation will evolve.

So, 2 questions:

1) Am I correct in thinking that it is better to use the same time lengths between the two for visualization purposes?

2) Is there any statistical way to 'standardize' the results (in terms of calculating percentages) of time series with different lengths?



You're intuition is right in my opinion.

1: The same length on the graph should correspond to the same amount of time but this doesn't stop one line from being longer than another line.

2: As for the growth rates: I think you would do better to measure them on a yearly basis so you compare apples and apples.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.