A note on Musgrave asymmetrical trend-cycle filters

Quenneville, Benoît; Ladiray, Dominique; Lefrançois, Bernard

doi:10.1016/s0169-2070(02)00080-8

“…This class generalizes the well-known Musgrave's asymmetric approximation of the Henderson filters [Musgrave (1964), see also Doherty (2001), Gray and Thomson (2002) and Quenneville, Ladiray and Lefrancois (2003)], which is implemented in the seasonal adjustment filter X-11, developed by the US Census Bureau [see Findley et al (1998) and Ladiray and Quenneville (2001)]. The class of filters depends on the properties of the true underlying signal, namely, its level, slope, curvature and so forth, which can be estimated from the data.…”

mentioning

confidence: 66%

Real Time Estimation in Local Polynomial Regression, with Application to Trend-Cycle Analysis

Proietti

¹

,

Luati

²

2008

View full text Add to dashboard Cite

The paper focuses on the adaptation of local polynomial filters at the end of the sample period. We show that for real time estimation of signals (i.e., exactly at the boundary of the time support) we cannot rely on the automatic adaptation of the local polynomial smoothers, since the direct real time filter turns out to be strongly localized, and thereby yields extremely volatile estimates. As an alternative, we evaluate a general family of asymmetric filters that minimizes the mean square revision error subject to polynomial reproduction constraints; in the case of the Henderson filter it nests the well-known Musgrave's surrogate filters. The class of filters depends on unknown features of the series such as the slope and the curvature of the underlying signal, which can be estimated from the data. Several empirical examples illustrate the effectiveness of our proposal.

show abstract

“…These techniques are by no means exhaustive of those available that have this property. For example, Bell and Martin (2004) provide a method for estimating the trend when the underlying unobserved components follow ARIMA processes, while Gray and Thomson (2002) and Quenneville et al (2003) extend the Henderson-Musgrave-type trend filters used in seasonal adjustment procedures such as X-12-ARIMA (Findlay et al, 1998) to enable them to be used right up to the end of the sample.…”

Section: Discussion and Summarymentioning

confidence: 99%

Modelling current trends in Northern Hemisphere temperatures

Mills

¹

2006

Intl Journal of Climatology

View full text Add to dashboard Cite

Fitting a trend is of interest in many disciplines, but it is of particular importance in climatology, where estimating the current and recent trend in temperature is thought to provide a major indication of the presence of global warming. A range of ad hoc methods of trend fitting have been proposed, with little consensus as to the most appropriate techniques to use. The aim of this paper is to consider a range of trend extraction techniques, none of which require 'padding' out the series beyond the end of the available observations, and to use these to estimate the trend of annual mean Northern Hemisphere (NH) temperatures. A comparison of the trends estimated by these methods thus provides a robust indication of the likely range of current trend temperature increases and hence inform, in a timely quantitative fashion, arguments based on global temperature data concerning the nature and extent of global warming and climate change. For the complete sample 1856-2003, the trend is characterised as having long waves about an underlying increasing level. Since around 1970, all techniques display a pronounced warming trend. However, they also provide a range of trend functions so that extrapolation far into the future would be a hazardous exercise.

show abstract

“…Musgrave asymmetric filters. Musgrave's asymmetric filters [Musgrave (1964), Doherty (2001), Quenneville, Ladiray and Lefrancois (2003)] are obtained in the particular case when the original two-sided symmetric filter is the Henderson filter and U = i, Z = [−h, −h + 1, . .…”

Section: 2mentioning

confidence: 99%

Real time estimation in local polynomial regression, with application to trend-cycle analysis

Proietti¹,

Luati²

2008

View full text Add to dashboard Cite

The paper focuses on the adaptation of local polynomial filters at the end of the sample period. We show that for real time estimation of signals (i.e., exactly at the boundary of the time support) we cannot rely on the automatic adaptation of the local polynomial smoothers, since the direct real time filter turns out to be strongly localized, and thereby yields extremely volatile estimates. As an alternative, we evaluate a general family of asymmetric filters that minimizes the mean square revision error subject to polynomial reproduction constraints; in the case of the Henderson filter it nests the well-known Musgrave's surrogate filters. The class of filters depends on unknown features of the series such as the slope and the curvature of the underlying signal, which can be estimated from the data. Several empirical examples illustrate the effectiveness of our proposal

show abstract

A note on Musgrave asymmetrical trend-cycle filters

Cited by 17 publications

References 5 publications

Real Time Estimation in Local Polynomial Regression, with Application to Trend-Cycle Analysis

Real Time Estimation in Local Polynomial Regression, with Application to Trend-Cycle Analysis

Modelling current trends in Northern Hemisphere temperatures

Real time estimation in local polynomial regression, with application to trend-cycle analysis

Contact Info

Product

Resources

About