# upward/downward bias of negative variable

If I have a variable that, considering some omitted factor, should have fallen by a higher amount than when it is not there - would that be a downward bias? I.e. the decrease is not large enough, so if I think in absolute terms (corresponding to an increase not being large enough) I would call that a downward bias. Does that carry over here? I am thinking it does :-)

• I don't think the use of bias here is correct. What is the context? Jan 15, 2019 at 11:26
• It is in the context of the effect of a change in one variable on the other. So say I change variable A by 1%, that changes B by -0.6% percent for example. Now, there might be an effect of B back on A not taken into account, such that A would actually have changed to a 0.8% change instead of 1%. So in principle, the effect on B on -0.6% actually corresponds to a change in A of 0.8% rather than 1%. In that way, for a 1% change in A, B should actually change downwards by more than 0.6% Jan 15, 2019 at 12:39