The regression method 
Top Previous Next 
This is an old method used before the ready availability of computers to solve the Maximum Likelihood equations. This method assumes a constant probability of capture. Because graphical and regression methods have been so widely used and are still frequently applied, we have included one in the program. It is inferior to the maximum likelihood approach, and does not give valid estimates of the standard error of the estimated population size.
If the number caught on the ith occasion is plotted against the total catch up to occasion i1, a straight line may be fitted by linear regression (Zippin 1956). The point where this line cuts the xaxis gives the estimated population size. In the example below, it is 169.66 or 170.
