Awerage growth rate (3 years) - which method?

[ynos]

Hello!



I am intending to perform independent samples T Test, between two groups of companies in comparing average annual growth rate in years 2002, 2003 and 2004.



Which way to calculate average growth rate is most suitable, for performing T Test later?


Data:


sales 2002 = 50

sales 2003 = 40

sales 2004 = 60


1.

[(sales 2004/sales 2002)^0,5]-1 = 0,095445115 = 9,545%


2.


[(sales 2003/sales 2002)-1 = -0,2 = -20%


[(sales 2004/sales 2003)-1 = 0,5 = 50%


average growth = (-20% + 50%)/2 = 15%




Any other method to calculate awerage growth rate?



tnx

 

[BioStatMatt]

The best way to calculate an average growth rate is probably a good question for an economics forum. However, I can provide a solution to this problem without losing the information that you would by taking averages. This can be accomplished using a 2-way analysis of variance where the factors are group and year. A significant interaction term between these two factors would indicate a difference in growth rate between groups.

I know this may sound a little complicated if you have not heard of anything I have just mentioned. However, the process is really straight forward. You could get help with the specifics by talking to a statistician at your school (if you are a student) or a consultant statistician.

~Matt