This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: j = 1 , 2 , . . . . (f) Derive the autocorrelation for all all lags j = 1 , 2 , . . . . Plot the autocorrelation function. (g) State the condition for which the model is invertible. (h) Assuming that the model is invertible, write it as an AR ( ∞ ). 1 Table 1: 3. Consider the EViews output for the time series regression of the series log( CPINSA t /CPINSA t − 4 ) where CPINSA contains quarterly values of the Consumer Price Index from 1961:Q1 to 2008:Q1. (a) Brie ﬂ y explain the relevance of using the series log( CPINSA t /CPINSA t − 4 ) as opposed to log ( CPINSA t ). (b) Why does EViews list the sample as starting from 1961:Q3 instead of 1961:Q1? (c) Is this series stationary? How do you know? (d) Is this series invertible? How do you know? 2...
View
Full Document
 Fall '11
 HENDERSON
 Economics, Econometrics, Variance, Autocorrelation, Stationary process, Autocovariance, CP IN SAt

Click to edit the document details