In parametric approach is a two step process and first a function form is assumed. Whereas no assumption is made in the non-parametric approach. Parametric approach reduces the problem of estimating the function to a few parameters, which simplifies the problem. Where as in non-parametric approach we try to find an estimate of the function that is very close to the actual data points as possible.
The main advantage of the parametric approach is simplification of the problem from estimating the function to estimating a few set of parameters. In addition the number of observation required to estimate the function is less compared to non-parametric approach.
The disadvantages with parametric approach is there is a possibility that the model chosen to estimate f is different from the actual true form of the f and this result in poor predictions. Also, trying to avoid this by choosing more flexible models might result in over fitting and increase in errors.
In parametric
approach
is a
two step
process and
first
a
function
form
is assumed
.
Whereas
no assumption
is made
in the non-parametric
approach
. Parametric
approach
reduces
the problem of estimating the
function
to a few parameters, which simplifies the problem.
Where as
in non-parametric
approach
we try to find an estimate of the
function
that is
very
close to the actual data points as possible.
The main advantage of the parametric
approach
is simplification of the problem from estimating the
function
to estimating a
few set
of parameters. In
addition
the number of observation required to estimate the
function
is less compared to non-parametric approach.
The disadvantages with parametric
approach
is there is a possibility that the model chosen to estimate f is
different
from the actual true form of the f and this result in poor predictions.
Also
, trying to avoid this by choosing more flexible models might result in over fitting and increase in errors.