Hi All

Hopefully someone here can help me with this, it's been a long time since I studied all this and I've forgotten how to do it.

Overview:

I work in advertising analysis and we have a database of information which includes data on the size of each campaign (measured in TVRs) and it's corresponding Coverage (ie what percentage of the population saw the advertisment). So for example a campaign of 100 TVRs might be viewed by 55% of the population and a campaign of 200 TVRs might be viewed by 70%. I have this data for about 1000 campaigns, so there is no shortage of data.

Problem:

What I need is an equation that I can use to predict coverage level when I feed in a number of TVRs, and vice-versa.

I created a Logarithmic trendline on a graph in Excel and that seemed to fit, but I noticed that the coverage is not limited to 100% as I need it to be. The Excel graph will also produce an equation for me but I would like to be able to produce the equation myself without having to make the graph each time.

I have attached the data.

Can you explain how I would construct the necessary equation and calculate it's component parts from the data.

Thank you very much for your help

Nick