Suppose that you have a time series dataset on the monthly U.S. national unemployment rate for the past 40 years. Do you think that unemployment is better characterized as a moving average process, an autoregressive process, or some combination of the two? Is there any evidence you can think of to support your claim? Explain.