Search results for "Change-point"

Examining nonstationarity in the recruitment dynamics of fishes using Bayesian change point analysis

2017

Marine ecosystems can undergo regime shifts, which result in nonstationarity in the dynamics of the fish populations inhabiting them. The assumption of time-invariant parameters in stock–recruitment models can lead to severe errors when forecasting renewal ability of stocks that experience shifts in their recruitment dynamics. We present a novel method for fitting stock–recruitment models using the Bayesian online change point detection algorithm, which is able to cope with sudden changes in the model parameters. We validate our method using simulations and apply it to empirical data of four demersal fishes in the southern Gulf of St. Lawrence. We show that all of the stocks have experience…

Atlantic salmon survival at sea: temporal changes that lack regional synchrony

2022

Spatial and temporal synchrony in abundance or survival trends can be indicative of whether populations are affected by common environmental drivers. In Atlantic salmon (Salmo salar L.), return rates to natal rivers have generally been assumed to be affected primarily by shared oceanic conditions, leading to spatially synchronous trends in mortality. Here, we investigated the existence of parallel trends in salmon sea survival, using data on migrating smolts and returning adults from seven Canadian populations presumed to share feeding grounds. We analysed sea survival, using a Bayesian change-point model capable of detecting non-stationarity in time series data. Our results indicate that w…

Change-point estimation in piecewise constant regression models and extensions

2015

Semiautomatic Behavioral Change-Point Detection: A Case Study Analyzing Children Interactions With a Social Agent

2021

The study of human behaviors in cognitive sciences provides clues to understand and describe people’s personal and interpersonal functioning. In particular, the temporal analysis of behavioral dynamics can be a powerful tool to reveal events, correlations and causalities but also to discover abnormal behaviors. However, the annotation of these dynamics can be expensive in terms of temporal and human resources. To tackle this challenge, this paper proposes a methodology to semi-automatically annotate behavioral data. Behavioral dynamics can be expressed as sequences of simple dynamical processes: transitions between such processes are generally known as change-points. This paper describes th…

Evaluating the performance of a new picking algorithm based on the variance piecewise constant models

2021

In this paper, a new picking algorithm for the automatic seismogram onset time determination is tested on a dataset of simulated waveforms. We aim at capturing the variations in the performance due to some characteristics of both the seismic event and its detection, which in turn affect some characteristics of the waveforms. We therefore simulate seismic events with different magnitude, and assumed to be detected with different distances from the nearest seismic station. Our tests permit to highlight the scenarios most suitable for our algorithm.

Is fiscal fatigue a threat to consolidation programmes?

2015

Building on a narrative approach to identify episodes of fiscal consolidation, data for a group of 17 industrial countries over the period 1978-2009 and continuous-time duration models, we find evidence suggesting that the likelihood of a fiscal consolidation ending increases over time, but only for programs that last less than six years. Additionally, fiscal consolidations tend to last longer in non-European than in European countries. Our results emphasize that chronic fiscal imbalances might lead to a vicious austerity cycle, while discipline in the behaviour of fiscal authorities is a means of achieving credible and shorter adjustment measures. Therefore, fiscal fatigue is likely to com…

The Legacy and the Tyranny of Time: Exit and Re-Entry of Sovereigns to International Capital Markets

2018

We use a novel continuous-time Weibull model (without and) with a change-point in the duration dependence parameter to investigate the duration of the exit and re-entry of sovereigns to international capital markets. Relying on annual data for a large panel of countries over the period 1970-2011, we find that, as the reputation of debtor countries as good (bad) borrowers solidifies over time, those episodes are more likely to end - i.e. the "legacy of time". Debtor countries can take advantage of the "benefit of doubt" of creditors during short exit spells. However, when exits are long and the reputation as a bad borrower emerges, no more "complacency" makes it more difficult for them to bo…

A fast and efficient picking algorithm for earthquake early warning application based on the variance piecewise constant models

2020

An earthquake warning system, or earthquake early warning system, is a system of accelerometers, seismometers, communication, computers, and alarms that is devised for notifying adjoining regions of a substantial earthquake while it is in progress. This is not the same as earthquake prediction, which is currently incapable of producing decisive event warnings. The implementation of efficient and computationally simple picking algorithm is necessary for this purpose, as well as automatic picking of seismic phases for seismic surveillance and routine earthquake location for fast hypocenter determination. In this paper a method for picking based on the detection of signals changes in variance …

Booms, Busts and normal times in the housing market

2015

We assess the existence of duration dependence in the likelihood of an end in housing booms, busts, and normal times. Using data for 20 industrial countries and a continuous-time Weibull duration model, we find evidence of positive duration dependence suggesting that housing market cycles have become longer over the last decades. Then, we extend the baseline Weibull model and allow for the presence of a change-point in the duration dependence parameter.We show that positive duration dependence is present in booms and busts that last less than 26 quarters, but that does not seem to be the case for longer phases of the housing market cycle. For normal times, no evidence of change-points is fo…

Change-points detection for variance piecewise constant models

2011

A new approach based on the fit of a generalized linear regression model is introduced for detecting change-points in the variance of heteroscedastic Gaussian variables, with piecewise constant variance function. This approach overcome some limitations of both exact and approximate well-known methods that are based on successive application of search and tend to overestimate the real number of changes in the variance of the series. The proposed method just requires the computation of a gamma GLM with log-link, resulting in a very efficient algorithm even with large sample size and many change points to be estimated.