Blue box example 10 - CFAi time series.
i’m confused with what they are considering a random walk. in the solution they say:
fx rates is a random walk because the estimated intercept does not appear to be significantly different than 0 and the estimated coefficient on the first lag is 0.9910 (close to 1).
so in my head this means that a random walk is define as b0 = 0 and b1 = 1
but then the following paragraph says:
is the exchange rate is a random walk, then b0=0 and b1= 0 and the error term will not be serially correlated.
aren’t these contradicting each other?
If they wrote _b_1 = 0 for a random walk, either it was a typo or the author hasn’t a clue; I’d lay money on the former.
In a random walk without drift, _b_0 = 0 and _b_1 = 1; in a random walk with drift, _b_0 ≠ 0 and _b_1 = 1.
if bo= 0.14 and b1=.9901
can you say with out calculating any t-statistic on b0 that it is a random walk?
_b_0 doesn’t matter in determining whether or not it’s a random walk, and 0.9901’s pretty darned close to 1. I’d stick my neck out and say that it is.
b_0 doesn’t matter in determining whether or not it’s a random walk, and 0.99_01’s pretty darned close to 1. 0.99_ 10 _'s even closer. I’d stick my neck out and say that it is.