# Question: Suppose an autoregressive model is used for data in which

Suppose an autoregressive model is used for data in which quarterly sales in 2013 were: 1.9, 1.7, 2.2, and 2.3 ($ Billion).

a) If a first-order autoregressive model is developed with estimated parameters of b0 = 0.100 and b1 = 1.12, compute the forecast for Q1 of 2014.

b) Compare this forecast to the actual value ($ 2.9B) by computing the absolute percentage error (APE). Did you over-forecast or under- forecast?

c) Assuming these quarterly sales have a seasonal component of length 4, use the following model to compute a forecast for Q4 of 2014: yt = 0.410 + 1.35 yt - 4. In fact, Q4 sales were $ 3.4B. Compare the APE for this forecast to that in part a. Compare the appropriateness of the different models.

a) If a first-order autoregressive model is developed with estimated parameters of b0 = 0.100 and b1 = 1.12, compute the forecast for Q1 of 2014.

b) Compare this forecast to the actual value ($ 2.9B) by computing the absolute percentage error (APE). Did you over-forecast or under- forecast?

c) Assuming these quarterly sales have a seasonal component of length 4, use the following model to compute a forecast for Q4 of 2014: yt = 0.410 + 1.35 yt - 4. In fact, Q4 sales were $ 3.4B. Compare the APE for this forecast to that in part a. Compare the appropriateness of the different models.

**View Solution:**## Answer to relevant Questions

Suppose an autoregressive model is used to model sales for a company that peaks twice per year (in June and December). a) What lagged variables would you try in a regression to forecast sales? Explain. b) How would you ...Average annual interest rates (banks prime lending) in the United States from 1966 through 2009 are shown in the following time series graph. a) What components do you see in this series? Here’s an autoregressive model ...In Exercise 39 we examined domestic tourists who visit Hawaii. Now, let’s consider international tourism. Here’s a time series plot of international visitors for the same time period. Here’s the linear trend model with ...A time series plot of monthly crude oil price ($/ barrel) from January 2001 to March 2007 is shown here. Using these data, a) Fit a first-order autoregressive model. b) Obtain a forecast for March 2007. For the Gas prices of Exercise 6, the actual value for January 2007 was 2.321. Find the absolute percentage error of your forecast.Post your question