Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
221 views2 pages

Ef3451 HW1 (Feb14 12)

This document contains instructions for a homework assignment on economic and business forecasting. It includes 4 questions related to time series analysis and forecasting models: 1) Interpreting regression coefficients and diagnostics from an OLS regression model. 2) Identifying which forecasting model diagnostic statistics are "better" when larger or smaller. 3) Forecasting future values using integrated ARIMA and AR models. 4) Deriving properties of the AR(1) model, including the mean, variance, and stationarity.

Uploaded by

Wayne WaiLok Chan
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
221 views2 pages

Ef3451 HW1 (Feb14 12)

This document contains instructions for a homework assignment on economic and business forecasting. It includes 4 questions related to time series analysis and forecasting models: 1) Interpreting regression coefficients and diagnostics from an OLS regression model. 2) Identifying which forecasting model diagnostic statistics are "better" when larger or smaller. 3) Forecasting future values using integrated ARIMA and AR models. 4) Deriving properties of the AR(1) model, including the mean, variance, and stationarity.

Uploaded by

Wayne WaiLok Chan
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 2

EF3451 Economic and Business Forecasting Homework assignment 1 Due on or before February 24, 2012 (Friday), 5: 30pm

1.

Using 30 annual observations on time series data, a forecaster obtained the following regression using the ordinary least squares method:

ln Yt =1.31 0.27 ln X 1t +0.53 ln X 2t 0.82 ln X 3t

(0.17)
R 2 = 0.615

(0.21)

(0.30)

, DW = 0.496 ,

where figures in brackets below parameter estimates are the corresponding estimated standard errors; Yt= number of business failures; X1t = volume of industrial production; X2t = short term interest rate; X3t = value of new business orders placed. Which of the following statements is true, which is false, and why? (a) The estimated partial regression coefficient on ln X 2t , 0.53, is interpreted as: all else remaining the same, one unit change in the short term interest rate leads to an expected 0.53 increase in business failures. (b) The value of R 2 will decline considerably if X 2t or X 3t is dropped. (c) The equation suffers from spurious regression problem. 2. (Source: Diebold, 2007, p.31, Exercise#2.2) For each of the diagnostic statistics listed here, indicate whether, other things the same, bigger is better, smaller is better, or neither. Explain your reasoning. (Hint: Be careful, think before you answer, and be sure to qualify your answers as appropriate.) (a) t-statistic (b) probability value (p-value) of the t-statistic (c) Standard error of the regression (d) Durbin-Watson statistic 3. (a) If an integrated ARIMA(0,1,0) model is used on the following data, what is the forecast for period t = 6 ? 1 2 3 4 5 6 Xt 10 12 13 15 17 ? (b) If an AR(1) model X t = 4 + 0.8 X t 1 is used on the following data, what is the forecast for t = 6 ? t 1 2 3 4 5 6 Xt 10 12 13 15 17 ? 4. For the AR(1) model
yt = yt 1 + t
t =1,2,..., T ;
2 with | |<1 and t iiN (0, )

Show that by successive substitution yt = t y0 + t 11 + ... + t . 2 2 (b) Show that if y 0 N (0, / 1 ) , then E ( yt ) = 0 for all t and 2 var( yt ) = 2 = /(1 2 ) so that the mean and variance are independent of t. a( yt ) = . If | |>1 then (c) Show that if =1 then v r v r yt ) is negative! a(
(a)

You might also like