Suppose the true model of return $r_t$ is

$r_t=+{}_1{f}_{1,t}+{}_2{f}_{2,t}+{}_t$

where $f_{1,t}$ and $f_{2,t}$ are two factors, and ${}_t$ is white noise,

a$)$ Econometrician Mr$.$ Omit is unaware of existence of factor $f_{2,t}$ and runs the following regression

$r_t=+{}_1{f}_{1,t}+{}_t$

Is hat$({)}_1$ from this regression $(2)$ an unbiased estimator of ${}_1$ in model $(1)?$ Is it consistent? If there is any biased, what is the direction of the bias?