Nüve Forum

Nüve Forum > akademik > İktisadi ve İdari Bilimler Fakültesi > Siyaset Bilimi ve Kamu Yönetimi Bölümü > How Important Are Structural Changes As An Explanation Of Unit Roots In Macroeconomic Time Series?

Siyaset Bilimi ve Kamu Yönetimi Bölümü hakkinda How Important Are Structural Changes As An Explanation Of Unit Roots In Macroeconomic Time Series? ile ilgili bilgiler


How Important Are Structural Changes As An Explanation Of Unit Roots In Macroeconomic Time Series? Some Counter Evidence From G-7 Countries In recent years there has been an increased concern

Like Tree1Likes
  • 1 Post By world

Cevapla

 

LinkBack Seçenekler Stil
  #1  
Alt 22.10.10, 15:04
world - ait kullanıcı resmi (Avatar)
Çılgın
 
Üyelik tarihi: Feb 2007
İletiler: 911
world artık çok görkemli biri.world artık çok görkemli biri.world artık çok görkemli biri.world artık çok görkemli biri.world artık çok görkemli biri.world artık çok görkemli biri.world artık çok görkemli biri.world artık çok görkemli biri.world artık çok görkemli biri.world artık çok görkemli biri.world artık çok görkemli biri.
Standart How Important Are Structural Changes As An Explanation Of Unit Roots In Macroeconomic Time Series?

How Important Are Structural Changes As An Explanation Of Unit Roots In Macroeconomic Time Series? Some Counter Evidence From G-7 Countries
In recent years there has been an increased concern about the effects of structural changes on the results obtained from the Dickey-Fuller unit root tests. Theoretically structural changes can cause artificial unit roots. Empirical validity of this result is, however, seen to be suspicious. Some studies claimed that the Dickey-Fuller tests reject the null hypothesis of a unit root, if a correction is made for the effects of structural changes. This study takes a different approach to examine the empirical validity of structural change effect. The study uses a split sample approach and applies some unit root tests with good power properties to the parts of sample that are believed to be not containing structural changes. Using several macroeconomic time series data from five G-7 countries, the study finds that subsamples of the data also contain unit roots. This result calls into question the structural change explanation of the presence of unit roots in most macroeconomic time series.
business-mix-(11).jpg
1. INTRODUCTION
Over the last decade, a great deal of literature has accumulated on the issue of unit roots in macroeconomic time series. Nelson and Plosser (1982) found that most of the US macroeconomic time series has a unit root, hence, is difference stationary (DS), or integrated of order one, 7(1), rather than trend stationary (TS), or integrated of order zero, 7(0), using the method developed in Dickey (1976), Fuller (1976), and Dickey and Fuller (1979, 1981) to test for unit roots. Time series that contain an autoregressive unit root are traditionally called nonstationary. Findings of Nelson and Plosser (1982) have been confirmed by other researchers over the years. However, the nonstationarity of the US real GNP series brought into question and there seems to be an opposite view developing about this series (see, for instance, Campbell and Mankiw (1987) and McCallum (1993)). In the following years after the publication of a series of papers by Dickey and Fuller, some new tests have been developed and some modifications and extensions have been suggested to the Dickey-Fuller unit root tests. These include Evans and Savin (1981, 1984), Sargan and Bahargava (1983), Said and Dickey (1984), Bahargava (1986), Dickey and Pantula (1987), Phillips (1987a), Phillips and Perron (1988), Perron (1989), Dufour and King (1991), Hall (1992), Kwiatkowski, Phillips, Schmidt, and Shin (1992), Elliot, Rothenberg, and Stock (1992), Schmidt and Phillips (1992), and Saikkonen and Luukkonen (1993) among many others.
Unit roots in macroeconomic time series has very different implications for the statistical analysis of these series and economic theorizing than a TS class macroeconomic time series. Engle and Granger (1987) showed a number of statistical properties specific to 7(1) series. The most important difference between 7(0) and 7(1) time series in terms of the statistical properties they possess is that the moments of the 7(1) series are time dependent while the moments of the 7(0) series are time invariant. To show that, let xt be a random walk without drift, then, it can easily be verified that for a
fixed starting value x0, xt has the representation xt = x0 + ^ ei with E(e2) = of and
var(xt) = to2, where et is a white noise. It is clear that the variance of a nonstationary time series will grow linearly with time. This is in sharp contrast with stationary time series whose moments are time invariant.

Mehmet BALCILAR
Assistant Professor, Department of Economics, Çukurova University, Faculty of Economics and Administrative Sciences, Adana
Eklenmiş Dosya
Dosya tipi: pdf balcilarhow.pdf (152,0 KB (Kilobyte), 13x kez indirilmiştir)
__________________
[CENTER][URL="http://www.nevart.net/"][IMG]http://www.nuveforum.net/galeri/data/500/2602.jpg[/img][/url]
Güzel Sanatlar Fakültesi/Lisesi Yetenek Sınavlarına Hazırlık Kursu
Resim Yağlı Boya Hobi Kursu
Hızlı ve Etkili Okuma Kursu
Çocuklar için Hızlı Okuma Kursu
Çocuklar için Resim Kursu
Diksiyon Kursu
Nefes Teknikleri Kursu
Kişisel Gelişim Kursları[/CENTER]
Alıntı ile Cevapla
Cevapla

Tags
explanation, important, macroeconomic, roots, series?, structural, time, unit

Seçenekler
Stil

Yetkileriniz
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is Açık
[IMG] Kodları Açık
HTML-KodlarıKapalı
Trackbacks are Açık
Pingbacks are Açık
Refbacks are Açık



Bütün zaman ayarları WEZ +2 olarak düzenlenmiştir. Şu anki saat: 07:59 .