# Question on Causation

• Jun 20th 2013, 04:53 AM
markviduka
Question on Causation
Hi Guys,

I am having some trouble with the following: supposed there is a simple multivariance equation Y = a +bx + bz. Y describes the dependent variable, x and z are independent variables. If it is possible that x and z effect y, is there anyway to control for this?

Kind Regards,
Mark
• Jun 20th 2013, 05:15 AM
HallsofIvy
Re: Question on Causation
I don't know what you mean by "control for this". Of course, the partial derivatives, in this simple linear case, just the constant b for both (if you had "y= a+ bx + cz" then b and c) measure how that variable affects y. If, by "control", you want to reduce or increase the influence of x and z on y (since y= a +b(x+ z), it really the influence of x+ z) just reduce or increase b. In the more general case, y= a+ bx+ cz, you can reduce or increase the influence of x on y by reducing or increasing b and the influence of z on y by reducing or increasing c.
• Jun 20th 2013, 05:52 AM
markviduka
Re: Question on Causation
Sorry. What i meant to say was: in the regression, x and z effect y. however, what if y effects x and z, is there anyway to control for this? So the direction of the relationship..