1. Suppose it is known that {Yt, te Z} is an AR(1) model with o approximately 0.7. How many observations of the time series do we need if we are to construct a 95% confidence interval on o whose width is no more than 0.1? Recall that for a E (0, 1), a 100(1 - a)% confidence interval on o is given by the interval (9 - za/2\\ Var($), $ + za/2 \\ Var($)), where Var($) = 1 - 62 n n is the number of observations, and zo/2 is the usual z-critical value, that is, zo/2 is such that the area under the standard normal distribution to the right z/2 is a/2