Change Point Detection Analysis Research Articles

Homogeneity and Change Point Detection Analysis of Seasonal and Annual Precipitation and Temperature Series Van, Türkiye

The quality and consistency of historical temperature and precipitation records are extremely important to researchers who study water resources, hydrological processes, and climate change. In this regard, Homogeneity tests are helpful tools for managing the accuracy and consistency of the data. In this study, the homogeneity of long-term annual and seasonal precipitation and temperature records obtained from five meteorological stations located in Van, Turkey, is examined using Standard Normal Homogeneity (SNHT), Pettitt (PT), Buishand Range (BR), and Cumulative Deviation (CD) tests at a significance level of 0.05. Finally, change-points were determined for each station where the homogeneity was disturbed. As a result of the study, there is mostly homogeneity in the precipitation data, and the homogeneity in the temperature data is deteriorated. The results of this study constitute a source of information in terms of the reliability of the meteorological data series. As a result, the reliability of the data should be questioned in the hydrological studies to be carried out in the Van region and the data should be made reliable in the projects to be carried out.

Trend detection and change point analysis of inflows in Karuppanadhi and Gundar dams of Chittar River Basin, Tamil Nadu, India

Abstract Gaining insights into the patterns of hydroclimatic elements such as inflows to dams is essential to adeptly strategize and oversee water resource planning. For this study, we focused on Karuppanadhi and Gundar dams within the Chittar River Basin in Tamil Nadu. We employed various time series trend assessment methods – Mann–Kendall (MK), modified Mann–Kendall (MMK), Sen slope estimator, and innovative trend analysis (ITA). Furthermore, we conducted change point detection analysis using homogeneity tests, namely Pettitt's test, standard normal homogeneity test (SNHT), and Buishand test. The analysis was carried out across three-time scales: monthly, seasonal, and annual for 30 years (1991–2020) span. Results revealed declining trends significantly across all three timescales for the Karuppanadhi dam. Whereas for the Gundar dam, notable trends included increased patterns in January, March, December, and the winter season, while other months/seasons showed statistically decreasing trends. ITA exhibited greater sensitivity in identifying trends at monthly and seasonal scales, indicating its superior trend detection over MK and MMK methods. Change point analysis of trends identified a rising trend in August post-2017 for both dams. Other change points indicated decreasing inflow trends thereafter. Analysing trends and change points in dam inflows aid decisions on water resources management.

Bayesian Analysis of Change Point Problems Using Conditionally Specified Priors

AbstractIn data analysis, change point problems correspond to abrupt changes in stochastic mechanisms generating data. The detection of change points is a relevant problem in the analysis and prediction of time series. In this paper, we consider a class of conjugate prior distributions obtained from conditional specification methodology for solving this problem. We illustrate the application of such distributions in Bayesian change point detection analysis with Poisson processes. We obtain the posterior distribution of model parameters using general bivariate distribution with gamma conditionals. Simulation from the posterior are readily implemented using a Gibbs sampling algorithm. The Gibbs sampling is implemented even when using conditional densities that are incompatible or only compatible with an improper joint density. The application of such methods will be demonstrated using examples of simulated and real data.

Homogeneity and change point detection of hydroclimatic variables: A case study of the Ghba River Subbasin, Ethiopia

In most studies on hydroclimatic variability and trend, the notion of change point detection analysis of time series data has not been considered. Understanding the system is crucial for managing water resources sustainably in the future since it denotes a change in the status quo. If this happened, it is difficult to distinguish the time series data’s rising or falling tendencies in various areas when we look at the trend analysis alone. This study’s primary goal was to describe, quantify, and confirm the homogeneity and change point detection of hydroclimatic variables, including mean annual, seasonal, and monthly rainfall, air temperature, and streamflow. The method was employed using the four-homogeneity test, i.e., Pettitt’s test, Buishand’s test, standard normal homogeneity test, and von Neumann ratio test at 5% significance level. In order to choose the homogenous stations, the test outputs were divided into three categories: “useful”, “doubtful”, and “suspect”. The results showed that most of the stations for annual rainfall and air temperature were homogenous. It is found that 68.8% and 56.2% of the air temperature and rainfall stations respectively, were classified as useful. Whereas, the streamflow stations were classified 100% as useful. Overall, the change point detection analyses timings were found at monthly, seasonal, and annual time scales. In the rainfall time series, no annual change points were detected. In the air temperature time series except at Edagahamus station, all stations experienced an increasing change point while the streamflow time series experienced a decreasing change point except at Agulai and Genfel hydro stations. While alterations in streamflow time series without a noticeable change in rainfall time series recommend the change is caused by variables besides rainfall. Most probably the observed abrupt alterations in streamflow could result from alterations in catchment characteristics like the subbasin’s land use and cover. These research findings offered important details on the homogeneity and change point detection of the research area’s air temperature, rainfall, and streamflow necessary for the planers, decision-makers, hydrologists, and engineers for a better water allocation strategy, impact assessment and trend analyses.

On stochastic dynamic modeling of incidence data.

In this paper, a Markov Regime Switching Model of Conditional Mean with covariates, is proposed and investigated for the analysis of incidence rate data. The components of the model are selected by both penalized likelihood techniques in conjunction with the Expectation Maximization algorithm, with the goal of achieving a high level of robustness regarding the modeling of dynamic behaviors of epidemiological data. In addition to statistical inference, Changepoint Detection Analysis is performed for the selection of the number of regimes, which reduces the complexity associated with Likelihood Ratio Tests. Within this framework, a three-phase procedure for modeling incidence data is proposed and tested via real and simulated data.

Spatio-temporal analysis of water requirement of maize (Zea mays L.) in Haryana state of India

Climate change have a considerable impact on crop water demand (ETc). The present study analyzed the spatio-temporal variation of the crop water requirement (ETc), crop water surplus deficit index (CWSDI), and the coupling degree of ETc and effective precipitation (Pe) for maize (Zea mays L.) crop grown in Haryana State of India using ArcMap 10.8 software. ETc was calculated using the Penman-Monteith method and crop coefficient (Kc) approach for 34 years (1985-2018) climatic data. Two statistical models, namely Mann–Kendall test and Sen’s slope estimator, were applied to understand the trend, while Pettitt's test was used to identify abrupt change points by using XLSTAT 2021 software.The result showed that for maize ETc ranged from 571 to 766 mm, CWSDI ranged from -29 to -74% and the degree of coupling of ETc and Pe ranged from 0.26 to 0.71 during 1985-2018. The decadal temporal variation of these indices indicated that ETc during 2007-2018 was relatively higher than it was from 1985-1995 and 1996-2006. The spatio-temporal values of CWSDI indicated that almost all the districts were under water deficit conditions for growing maize crop.Trend analysis showed significantly increasing trend of CWR for most of the districts except Jind, Mewat, Palwal, Panchkula, Rewari, Rohtak and Yamunanagar.The abrupt change point detection analysis of CWR indicates toward different change points, with maximum change points between the years 2004–2008.Such information will help to the farmers in the effective management of water for sustainable production under climate change in the near future.

COVID-19’s Impact on Global Stock Market Movements and an Examination on Turkish Market

Finansal küreselleşme, 1980’li yıllarda gelişmiş ülkelerde ve 1990’lı yıllarda gelişmekte olan ülkelerde ortaya çıkan sermaye hareket serbestlikleri ile birlikte önemli bir aşamaya geçmiştir. Bu aşamada sistemik risk unsurları artmış, olumsuz gelişme ve beklentiler tüm piyasalara hızla yayılarak etki edebilir hale gelmiştir. Hisse senedi piyasaları özelinde bakıldığında çok sayıda ülke piyasasının “küresel” piyasaya hızlı bir şekilde entegre olduğu görülmüş ancak söz konusu entegrasyonun düzeyi, tüm ülke piyasalarının güçlü bir şekilde birlikte hareket edecekleri seviyelere ulaşmamıştır. Yaşanan krizler karşısında kısa dönemde benzer reaksiyonlar gözleniyor olsa da şoklar sonrası hareketlilikler önemli farklılıklar gösterebilmektedir. Bu çerçevede, COVID-19 hastalığının Dünya Sağlık Örgütü tarafından 11 Mart 2020 tarihinde küresel ölçekte salgın ilan edilmesi ile hisse senedi piyasalarının birlikte hareketlerinde bir değişim olup olmadığı, olduysa değişimin ne şekilde olduğu temel bileşenler analizi ve değişim noktası analizi ile araştırılmıştır. Analizde 31.12.2020 tarihine kadar gerçekleşen günlük endeks hareketleri kullanılmıştır ve Türkiye ile birlikte 20 ülke piyasası dâhil edilmiştir. Ayrıca Türkiye hisse senedi piyasasının, ülkede salgının seyrine ilişkin açıklanan sayı ve kısıtlamalardan ne düzeyde etkilendiği de incelenmiştir.

Introducing change point detection analysis in relationship research: An investigation of couples’ emotion dynamics

Many relationship theories assume some form of interdependence between relationship partners. Partners are thought to continuously influence each other and to be influenced by each other over time. These influences are not expected to be constant, but dynamic (sometimes partners influence each other a lot, and sometimes they do not influence each other). To investigate such changes in interpersonal dynamics, we showcase the value of using a change point detection approach, which can be used to monitor virtually any preferred quantification of interpersonal dynamics across time. Concretely, we introduce the KCP-RS method, which scans times series for changes in user-specified statistics, in interpersonal emotion dynamic research. We used KCP-RS to investigate changes in 96 couples’ emotional experiences during two 10-minute conversations, which were meant to elicit a negative and a positive interaction context. Based on participants’ continuous reports of the valence of their emotional experience, we looked for changes in three statistical measures, aiming to capture emotional similarity between partners (i.e., does their valence fluctuate together). Specifically, we investigated the occurrence, frequency, and direction of change in partners’ linear correlations, instantaneous derivative matching (IMD), and signal matching (SM). While correlation changes were only observed in 2% of the couples, IDM changes were detected for about one third of the couples (34%), and SM changes were detected in about half of them (49%). Most couples demonstrated one change point, and the direction of the change differed depending on the specific emotional similarity measure. In a first validation of this method, we demonstrated how such change points can pinpoint to subtle but meaningful dynamic processes in couples. We end by discussing the added value of change point detection analyses for relationship research and interpersonal research in general.

Recent changes in the climate-growth response of European larch (Larix decidua Mill.) in the Polish Sudetes

Key messageRecently observed temporal changes in the climate-growth relationship of larch in the Polish Sudetes suggest growth limitations in the future.Larches in the Sudetes are very sensitive to the currently changing climatic factors, and an extreme negative response to drought is observed. In this study, temporal changes in the climate-growth relationship of European larch were analyzed using moving-window correlation. Change-point detection analysis was performed to determine whether there is a temporal connection between tree-ring growth responses and changes in climatic factors trends. The Random Forest predictor importance determination method was used to establish the set of climatic factors that influence larch tree-ring growth the most and to show how this set changes over time. Additionally, cluster analysis was applied to find spatial growth patterns and to generalize the growth response of larch. The results indicate that the main clustering factor is altitude. Nevertheless, an increasing unification of the larch’s response to dominant climatic factors is observable throughout the whole study area. This unification is expressed in the increasingly positive and recently dominant effect of May temperature. A progressively negative influence of the temperature in the summer and late autumn of the year preceding growth was observed, as was an increasing influence of water availability in the summer months. The study indicates that there is a connection between the observed changes and the recent rapid rise in temperature, which has consequently had a negative influence on water availability. The growth of this tree species in the Sudetes is expected to be very limited in the future due to its sensitivity to drought, the predicted increase in temperatures and thermal extremes, and the decrease of the share of summer precipitation in the annual total.

Flyway-scale analysis reveals that the timing of migration in wading birds is becoming later.

Understanding the implications of climate change for migratory animals is paramount for establishing how best to conserve them. A large body of evidence suggests that birds are migrating earlier in response to rising temperatures, but many studies focus on single populations of model species.Migratory patterns at large spatial scales may differ from those occurring in single populations, for example because of individuals dispersing outside of study areas. Furthermore, understanding phenological trends across species is vital because we need a holistic understanding of how climate change affects wildlife, especially as rates of temperature change vary globally.The life cycles of migratory wading birds cover vast latitudinal gradients, making them particularly susceptible to climate change and, therefore, ideal model organisms for understanding its effects. Here, we implement a novel application of changepoint detection analysis to investigate changes in the timing of migration in waders at a flyway scale using a thirteen‐year citizen science dataset (eBird) and determine the influence of changes in weather conditions on large‐scale migratory patterns.In contrast to most previous research, our results suggest that migration is getting later in both spring and autumn. We show that rates of change were faster in spring than autumn in both the Afro‐Palearctic and Nearctic flyways, but that weather conditions in autumn, not in spring, predicted temporal changes in the corresponding season. Birds migrated earlier in autumn when temperatures increased rapidly, and later with increasing headwinds.One possible explanation for our results is that migration is becoming later due to northward range shifts, which means that a higher proportion of birds travel greater distances and therefore take longer to reach their destinations. Our findings underline the importance of considering spatial scale when investigating changes in the phenology of migratory bird species.

Frequency, timing and risk factors for primary maternal cytomegalovirus infection during pregnancy in Quebec.

Maternal Cytomegalovirus (CMV) infection in the first trimester (T1) of pregnancy is a public health concern, as it increases the risk of severe neurodevelopmental outcomes associated with congenital infection compared to infections occurring later during pregnancy. To determine CMV seroprevalence in T1 of pregnancy, its trend, risk factors and the incidence rate of primary infection during pregnancy. Using the biobank of the prospective cohort "Grossesse en Santé de Québec" collected between April 2005 and March 2010 at the Québec-Laval Hospital, Québec, Canada, maternal CMV serology was determined using Abbott Architect Chemiluminescence microparticle immunoassays for immunoglobulin G(IgG), immunoglobulin M(IgM) titration and IgG avidity testing. Changepoint detection analysis was used to assess temporal trends. Risk factors associated with seropositivity were determined by multivariable logistic regression. CMV seroprevalence in T1 of pregnancy was 23.4% (965/4111, 95% CI, 22.1-24.7%). The incidence rate for CMV primary infection during pregnancy was 1.8 (95% CI, 1.2-2.6) per 100 person-years. No changepoint was identified in the maternal CMV-seroprevalence trend. Multivariable analyses showed that T1 maternal CMV seropositivity was associated with having one child OR 1.3 (95% CI, 1.10-1.73) or two or more children OR 1.5 (95%CI, 1.1-2.1), ethnicity other than Caucasian OR 2.1 (95% CI, 1.1-3.8) and country of birth other than Canada and the USA OR 2.8 (95% CI, 1.5-4.9). In this cohort, maternal seroprevalence in T1 of pregnancy and seroconversion rate were low. This information and identified risk factors could help guide the development and implementation of preventive actions and evidence-based health policies to prevent CMV infection during pregnancy.

Statistical analysis of modified Hargreaves equation for precise estimation of reference evapotranspiration

There are a lot of meteorological stations around the world, like Pakistan, where all the meteorological parameters are not accessible to analyze Penman-Monteith (PM) method for the estimation of Reference Evapotranspiration (ETo ). So ablative methods such as Hargreaves Samani (HS) equation are therefore a potential solution to this problem, which required small numbers of meteorological parameters to estimate ETo . But, simplification brings imprecision and the need for regional calibration of the HS equation’s constant. In this study, fuzzy logic based calibration of the constants in the HS equation at the various climatic locations of Pakistan for the period of 1980 − 2019 is performed. The precision of HS and Modified HS (MHS) against standard PM method for the approximation of ETo was evaluated and compared the calibration process of HS equation by adjusting the coefficients of the equation gradually. The performance of the MHS is evaluated based on the statistical indicators such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Performance Index (PI). The statistical assessment of the study revealed that the MHS equation performed better as compared to the conventional HS equation and simultaneous adjustment of all the empirical parameters “ah”, “bh”,”ch” and “dh” was the best alternative for calibration of the HS equation. Furthermore, the change point detection analysis has been carried out on ETo (estimated by MHS) using Pettitt’s test, Buishand’s range test and Buishand’s U test. Results of the change point analysis indicated different change points from year 1980 to 2019, mainly due to the industrialization and urbanization during this time period.

Periodic-type auto-regressive moving average modeling with covariates for time-series incidence data via changepoint detection.

When it comes to incidence data, most of the work on this field focuses on the modeling of nonextreme periods. Several attempts have been made and a variety of techniques are available to achieve so. In this work, in order to model not only the nonextreme periods but also capture the behavior of the whole time-series, we make use of a dataset on influenza-like illness rate for Greece, for the period 2014-2016. The identification of extreme periods is made possible via changepoint detection analysis and model selection techniques are developed in order to identify the optimal periodic-type auto-regressive moving average model with covariates that best describes the pattern of the time-series. In addition, in the context of incidence data modeling, an advanced algorithm was developed in order to improve the accuracy of the selected model. The derived results are satisfactory since the changepoint method seems to identify correctly the extreme periods, and the selected model: (1) estimates accurately the influenza-like illness syndrome morbidity burden in the case of Greece, and (2) captures satisfactorily enough the behavior of the whole time-series.

Analysis of Drivers of Trends in Groundwater Levels Under Rice–Wheat Ecosystem in Haryana, India

In India, especially in the state of Haryana in the northwestern parts, groundwater is the key source of water supply for domestic, industrial and agricultural uses. In recent years, the dependence on groundwater has increased drastically over the state due to speedy expansion of area under rice–wheat cultivation and therefore the states’ aquifers have started experiencing over-exploitation. The aim of this study was to compute the magnitudes of groundwater level trends and abrupt change point detections spatially and temporally based on secondary data obtained from the Groundwater Cell, Department of Agriculture, Government of Haryana with respect to 118 community development blocks for pre- and post-monsoon seasons during 1990–2013. Three statistical models, namely Mann–Kendall test, Sen’s slope estimator and simple linear regression, were applied to understand the trend and rate of change in groundwater level, whereas the abrupt change point detection analysis was carried out using Pettitt’s test, standard normal homogeneity test, Buishand’s range test and von Neumann ratio test. The results indicate that the average depth to groundwater level during pre-monsoon season in the state has ranged from 3.42 to 43.90 m below ground level (mbgl), whereas the average depth to groundwater level during post-monsoon season has ranged from 2.48 to 43.68 mbgl. The mean of groundwater level fluctuations during pre- and post-monsoon seasons was found to be − 6.0 and − 6.7 mbgl, respectively. The study highlights that the negative (falling) trends are much more pronounced (significant at 0.01% probability level) than positive (rising) ones, which corroborate the unsustainable groundwater development in the state. The abrupt change point detection analysis of groundwater levels indicates toward different change points, with maximum change points between the years 2000–2006 during pre- and post-monsoon seasons. This can be attributed to the advanced transplanting of rice crop by farmers much before the start of rainy season. In addition, both anthropogenic and climatic factors have led to groundwater depletion in the study area. These findings may be helpful for the planners and policy makers toward judicious utilization of groundwater resources in the state.

Trends in annual and extreme flows in the Lena River basin, Northern Eurasia

AbstractA long‐term daily streamflow data set, covering the period from 1925 to 2013, was compiled for the Lena River basin, with an average record length of 49 years. Time series of mean annual daily flow (MADF) and extreme (maximum, Qmax, and minimum, Qmin) daily flows were subject to trend detection and change point detection analysis. Significant changes result mostly from rapid “breakpoint” homogeneity disruptions. Thirty one time series showed trends in MADF, 10, in Qmax, and 32 of the 55 records with nonzero Qmin, significant at p ≤ 0.05. Upward trends prevail in both mean annual and extreme flows, with average magnitudes of 47% (MADF), 56% (Qmax), and 68% (Qmin). Two to three stations in each subset showed downward trends (averaging −20%, −39%, and −38%, respectively). Abrupt changes are observed in the 1990s and early 2000s mostly in the headwaters of the basin, underlain by discontinuous permafrost.

Application of singular spectrum-based change-point analysis to EMG-onset detection

While many approaches have been proposed to identify the signal onset in EMG recordings, there is no standardized method for performing this task. Here, we propose to use a change-point detection procedure based on singular spectrum analysis to determine the onset of EMG signals. This method is suitable for automated real-time implementation, can be applied directly to the raw signal, and does not require any prior knowledge of the EMG signal’s properties. The algorithm proposed by Moskvina and Zhigljavsky (2003) was applied to EMG segments recorded from wrist and trunk muscles. Wrist EMG data was collected from 9 Parkinson’s disease patients with and without tremor, while trunk EMG data was collected from 13 healthy able-bodied individuals. Along with the change-point detection analysis, two threshold-based onset detection methods were applied, as well as visual estimates of the EMG onset by trained practitioners. In the case of wrist EMG data without tremor, the change-point analysis showed comparable or superior frequency and quality of detection results, as compared to other automatic detection methods. In the case of wrist EMG data with tremor and trunk EMG data, performance suffered because other changes occurring in these signals caused larger changes in the detection statistic than the changes caused by the initial muscle activation, suggesting that additional criteria are needed to identify the onset from the detection statistic other than its magnitude alone. Once this issue is resolved, change-point detection should provide an effective EMG-onset detection method suitable for automated real-time implementation.