Most logical value

I would like to know if there is a method of determining "the most
logical" value in a table.
These are some data that I put in the table below. These numbers are
obtained from some dedicated software to compute the influence of an
investment on some projects. I would like to fill in the gaps myself.
I can add the data from column 10 en 20 and divide it by 2 to compute
the value for column 15, but is there a better method?

investment
year 0 5 10 15 20 25 30
2005 72,8 72,8 72,8 72,8 72,8 72,8 72,8
2006 79,4 76,7 71,5 60,9 50,3
2007 92,3 87,4 76,9 55,6 33,5
2008 108,6 102,3 86,2 53,3 19,0

Thanks,

Alex

That assumes a linear model. You can use linest to develop the coeficients
for a non-linear model.

You can see how Bernard Liengme solves for the coefficients of a polynomial
curve using LINEST.

http://www.stfx.ca/people/bliengme/E...Polynomial.htm


--
Regards,
Tom Ogilvy


"Alex" <[email protected]> wrote in message
news:[email protected]...
> I would like to know if there is a method of determining "the most
> logical" value in a table.
> These are some data that I put in the table below. These numbers are
> obtained from some dedicated software to compute the influence of an
> investment on some projects. I would like to fill in the gaps myself.
> I can add the data from column 10 en 20 and divide it by 2 to compute
> the value for column 15, but is there a better method?
>
> investment
> year 0 5 10 15 20 25 30
> 2005 72,8 72,8 72,8 72,8 72,8 72,8 72,8
> 2006 79,4 76,7 71,5 60,9 50,3
> 2007 92,3 87,4 76,9 55,6 33,5
> 2008 108,6 102,3 86,2 53,3 19,0
>
> Thanks,
>
> Alex

Tom Ogilvy wrote:
> That assumes a linear model. You can use linest to develop the coeficients
> for a non-linear model.
>
> You can see how Bernard Liengme solves for the coefficients of a polynomial
> curve using LINEST.
>
> http://www.stfx.ca/people/bliengme/E...Polynomial.htm
>
>
Thank you!
In my model it should be sufficient to use TREND. I found it when I was
looking at LINEST ;-)

Alex