Delta method

#1
I was wondering if anybody can help me as I've been stuck on this problem for ages and my attempts seem to all be incorrect so far.


I have four estimators (W,Χ,Y,,Z) and standard errors for each. How can the standard error of D be cacluated from the following equation, where K, and L are constants?

D=W(K/(XY(Z+L)))

I believe I need to use the delta method for this also but I have no idea how to go about it.
 
#3
Thank you. I have looked at almost every website I can find on the delta method and from what I can gather
VarXYZ=D^2.[(VarX/X^2)+(VarY/Y^2)+(VarZ/Z^2)]

But I can't work out how this changes when a constant divides XYZ.
 
#4
Thank you. I have looked at almost every website I can find on the delta method and from what I can gather
VarXYZ=D^2.[(VarX/X^2)+(VarY/Y^2)+(VarZ/Z^2)]

But I can't work out how this changes when a constant divides XYZ.
 

BGM

TS Contributor
#5
It is just following the (Multivariate) Taylor Expansion principle, expanded about the mean point. Do you find difficulties when following the wiki page?