a.m. say wages equal to be 20 plus beta1
education plus are care whatever
am if educations equal to 0
we want to know what are we drop the and our wage will be equal to
RBS euro 4
our beta1 its late the
slightly harder to interpret but its
nothing we can't handle easily so free take the derivative
love why in respect to
acts we will obtain
be the one
so this is what we will obtain
if we have a causal relationship so
om if we take the derivative this we can break this down
if we take the derivative this this is just a constant
so it will become 0 them X has no
exponentials in Seoul this whole term here will just become
the coefficient which is beta 1 and then if we have a causal relationship
we have to assume something called the conditional mean
independence assumption which assigns
essentially means the the derivative
love you in respect to axe
is equal to 0 am
in it sometimes also shown
since the expected value you
given X equal to 0 so
this just means that X
isn't for you is not dependent on so there's nothing
in this a return that
effects um acts
so if we were going to go back to are example
we had wage is
equal to some constant plus
paid online education
plus you
and we can not
assume that this is a causal relationship in
that does not mean that education does not affect wage because it
definitely to us course it does but what it means is that
their a.m. causal means that there's nothing
in you they is that
education is dependent on but we can sort of
a.m. user intuition to save it
you'll probably have things such as ability
and ability is generally
correlated very strongly with education
in because this is in this a return we from
we didn't include it in are regression um
then we we will not be able to assume the
conditional mean assumption and therefore these coefficients here
data 0 beta1 will not be completely correct
home they might be overestimated or underestimated depending upon
the the
omitted variables so omitted variables will get into