Modelling Autoregressive Processes with a Shifting Mean

González, Andrés; Teräsvirta, Timo

doi:10.2202/1558-3708.1459

Cited by 18 publications

(26 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For this reason we name our estimation strategy SlowShift, as opposed to González and Teräsvirta's (2008) QuickShift procedure. The advantage of SlowShift is that with a fine enough grid, the in-sample mean square prediction error is effectively minimized.…”

Section: Logistic Function Componentsmentioning

confidence: 99%

“…As González and Teräsvirta (2008) note, there is less need to have an evenly spaced grid for relatively large values of γ.…”

Section: Notesmentioning

confidence: 99%

“…In so doing we employ a set of new tools for examining smooth shifts in commodity prices. Specifically, we implement two variants of a smooth shifting-mean autoregressive (SM-AR) process: one due to Becker, Enders, and Hurn (2004;2006b) that is based on a Fourier flexible form; and one due to González and Teräsvirta (2008) that is based on a time-varying autoregressive (TV-AR) model. Because SM-AR models do not force structural change to be sharp they represent a reasonable alternative to the more common Bai and Perron (1998) approach, the results of which are also included for comparison.…”

Section: Introductionmentioning

confidence: 99%

“…In section three we describe the data, while section four reviews unit root testing in the presence of shifting means and reports the results of these tests applied to the commodity price data. In section five we detail the Bai and Perron (1998), the Fourier, and the González and Teräsvirta (2008) methodologies and compare the estimated break dates found by each. Section six discusses the implications of our analysis for the timing and possible causes of changing commodity prices.…”

Section: Introductionmentioning

confidence: 99%

“…González and Teräsvirta (2008) also assume the series in question moves around a deterministically shifting mean while allowing shifts to be either sharp or smooth. Specifically, they consider:…”

mentioning

confidence: 99%

See 4 more Smart Citations

Sharp Breaks or Smooth Shifts? an Investigation of the Evolution of Primary Commodity Prices

Enders

Holt

2012

American J Agri Economics

View full text Add to dashboard Cite

This paper explores the behavior of real commodity prices over a 50-year period. Attention is given to how the shifting means for various commodity prices have changed with a special emphasis on behavior since the mid 2000s. To identify structural changes in commodity prices, we estimate shifting-mean autoregressions by using: the Bai and Perron (1998) procedure for determining structural breaks; low frequency Fourier functions; and a procedure that specifies shifts to be smooth logistic functions of time. We find that the pattern in the timing of shifts is suggestive of the causal factors underlying the recent boom.

show abstract

Section: Logistic Function Componentsmentioning

confidence: 99%

“…As González and Teräsvirta (2008) note, there is less need to have an evenly spaced grid for relatively large values of γ.…”

Section: Notesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

Sharp Breaks or Smooth Shifts? an Investigation of the Evolution of Primary Commodity Prices

Enders

Holt

2012

American J Agri Economics

View full text Add to dashboard Cite

show abstract

Detecting Volcanic Eruptions in Temperature Reconstructions by Designed Break‐Indicator Saturation

Hendry¹,

Pretis²,

Schneider³

et al. 2017

Environmental Economics and Sustainability

View full text Add to dashboard Cite

We present a methodology for detecting structural breaks at any point in timeseries regression models using an indicator saturation approach. Building on recent developments in econometric model selection for more variables than observations, we saturate a regression model with a full set of designed break functions. By selecting over these break functions using an extended general-to-specific algorithm, we obtain unbiased estimates of the break date and magnitude. Monte Carlo simulations confirm the approximate properties of the approach. We assess the methodology by detecting volcanic eruptions in a time series of Northern Hemisphere mean temperature spanning roughly 1200 years, derived from a fully-coupled global climate model simulation. Our technique demonstrates that historic volcanic eruptions can be statistically detected without prior knowledge of their occurrence or magnitude-and hence may prove useful for estimating the past impact of volcanic events using proxy-reconstructions of hemispheric or global mean temperature, leading to an improved understanding of the effect of stratospheric aerosols on temperatures. The break detection procedure can be applied to evaluate policy impacts as well as act as a robust forecasting device. * Financial support from the Oxford Martin School, Open Society Foundations, Robertson Foundation, and British Academy, is gratefully acknowledged. We thank Max Roser,

show abstract