Project user data

I'm sure this is something simple that I'm overlooking. I really don't know much about Excel except what I've been forced to learn on the job.

I have sets of data with three kinds of data:
Column A is years
Column B is number of persons using a specific service.
Column C is number of persons within the total population set

I need to project out to 2015

Currently I'm using the Forecast function like so:
=FORECAST(2011,B3:B6,A3:A6)

The problem is that the data results that I'm getting don't seem to jive with expectations.. they seem a bit too inflated.

What would be a better function to use in order to project the change in the users and population based on the past trends?

Thanks so much for any help!

PS - This is my first post.. I spent a lot of time searching and could not find anything that made sense for my application so I apologize if I missed something.

PPS - I attached a file to show what I'm talking about. Sorry I didn't do it sooner.

You don't have your cell references locked into place (absolute cell references) thus when you drag your formula down, it's changing the cells it's looking at. In your example, in B7
=FORECAST(A7,B$3:B$6,$A$3:$A$6) can be dragged to the right and down. That lowers your numbers User numbers and raises your Population numbers somewhat. Did you try graphing to see what it looks like?
Also note, you are not using your initial values (row 2). Is that on purpose?

Originally Posted by ChemistB

You don't have your cell references locked into place (absolute cell references) thus when you drag your formula down, it's changing the cells it's looking at. In your example, in B7
=FORECAST(A7,B$3:B$6,$A$3:$A$6) can be dragged to the right and down. That lowers your numbers User numbers and raises your Population numbers somewhat. Did you try graphing to see what it looks like?

Yes, I have it graphed as well.

Thank you for the suggestion to lock down the cell data. I didn't know about how to do that and I was able to look it up based on your statements.

Originally Posted by ChemistB

Also note, you are not using your initial values (row 2). Is that on purpose?

I was told that because there is such a large information gap (from 2000 to 2005) that it would cause the resulting numbers to be flawed. If this is not the case then I'll use it.

Thanks for responding.

Another key function when looking at statistical forecasting like this is the correlation factor which defines how good your x and y values match. In Excel, it's
=CORREL(y-range,x-range)
A value of 1.000 is a perfect correlation.