= Adams–Moulton method
{c}
{wiki}
An <Adams–Moulton method> is an implicit <linear multistep method> obtained by interpolating derivative values including the new time level and integrating that <interpolation polynomial> over the next interval. For uniform step size $h$, linear interpolation through $f_n,f_{n+1}$ gives the second-order <trapezoidal rule>
$$
Y_{n+1}=Y_n+\frac h2(f_n+f_{n+1}).
$$
Quadratic interpolation through $f_{n-1},f_n,f_{n+1}$ instead gives the third-order formula
$$
Y_{n+1}=Y_n+\frac h{12}(5f_{n+1}+8f_n-f_{n-1}),\qquad f_j=f(t_j,Y_j).
$$
Because $f_{n+1}$ depends on the unknown $Y_{n+1}$, a nonlinear <vector field> generally requires a solve. Substituting a predicted new value only once defines a different composite numerical procedure; it should not automatically inherit the fully implicit method's <absolute stability> region.
Back to article page