Detection of Undocumented Changepoints: A Revision of the Two-Phase Regression Model

By Robert B. Lund, Jaxk Reeves. 2002

308 citations recorded by the source corpus.

Subfields: Applied Mathematics, Management Science and Operations Research, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Applied Mathematics

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Differentially Private Change-Point Detection (2018, background): “The change-point detection problem originally arose from industrial quality control, and has since been applied in a wide variety of other contexts including climatology [LR02], econometrics [BP03], and DNA analysis [ZS12].”
Penalized Maximal t Test for Detecting Undocumented Mean Change in Climate Data Series (2007, methodology): “In the climate literature, the most commonly used methods for changepoint detection include the standard normal homogeneity test (SNHT; Alexandersson 1986), the cumulative deviation test (Buishand 1982), two-phase regression-based methods (Solow 1987; Easterling and Peterson 1995; Lund and Reeves 20”
Exact and efficient Bayesian inference for multiple changepoint problems (2006, background): “…rate parameter (Raftery and Ak-man, 1986; Yang and Kuo, 2001; Ritov et al. , 2002), changing linear regression models (Carlin et al. , 1992; Lund and Reeves, 2002), Gaussian observations with varying mean (Worsley, 1979) or variance (Chen and Gupta, 1997; Johnson et al. , 2003), and Markov…”
New Techniques for the Detection and Adjustment of Shifts in Daily Precipitation Data Series (2010, methodology): “where S0 2 denotes the sum of residual squares of the null model (3) fit and S(2)(c) denotes the sum of the full model (1) fit with a changepoint at time tc; in addition, k is the number of free regression parameters under the full model (1) and kd is equal to the difference in the number of free re”
Penalized Maximal F Test for Detecting Undocumented Mean Shift without Trend Change (2008, methodology): “…F test (TPR3) for mean shifts in time series with a constant linear trend throughout the series (Wang 2003), and the two-phase regression model– based maximal F test for a sudden change that occurs simultaneously in both the mean and the linear trend of the series (TPR4; Lund and Reeves 2002).”
Use of change-point detection for friction–velocity threshold evaluation in eddy-covariance studies (2013, methodology): “An attractive but previously untested alternative to MPT is change-point detection (CPD), a technique that is widely used to detect temporal discontinuities in climatic data (Solow, 1987; Lund and Reeves, 2002; Wang, 2003).”
Stratospheric water vapor trends over Boulder, Colorado: Analysis of the 30 year Boulder record (2011, methodology): “This was accomplished using the revised two‐phase regression model of Lund and Reeves [2002] to identify undocumented changepoint dates in the time series for each 2 km layer.”
Comparison of two homogenized datasets of daily maximum/mean/minimum temperature in China during 1960–2013 (2016, methodology): “Li et al. (2014) compared the differences among four homogenized temperature series based on MASH, two-phase regression (Lund and Reeves, 2002), the Relative Homogenization test (RHtest) (Wang et al., 2007; Wang, 2008a), and HOM method for Henan Station in Qinghai Province.”
On‐line inference for multiple changepoint problems (2007, background): “Examples of changepoint problems include Poisson processes with changing intensity (Ritov et al., 2002), changing linear regressions (Lund and Reeves, 2002), Gaussian models with changing variance (Johnson et al., 2003) and Markov models with time-varying transition matrices (Braun and Muller,…”
Bayesian multivariate linear regression with application to change point models in hydrometeorological variables (2007, methodology): “Most of the published methodologies use classical statistical hypothesis testing to detect changes in slopes or intercept of linear regression models [Solow, 1987; Easterling and Peterson, 1995; Vincent, 1998; Lund and Reeves, 2002; Wang, 2003].”
Accounting for Autocorrelation in Detecting Mean Shifts in Climate Data Series Using the Penalized Maximal t or F Test (2008, methodology): “…the existing regression based methods for detecting type-1 changepoints in time series of IID Gaussian errors (i.e., Wang et al. 2007; Wang 2008; Lund and Reeves 2002; Alexandersson 1986) are also empirical approaches, although the distribution of the test statistic is independent of the time…”
A Review and Comparison of Changepoint Detection Techniques for Climate Data (2007, methodology): “The Wmax statistic or variants thereof have been proposed by several climate authors, most notably Karl and Williams (1987), with a refinement by DucréRobitaille et al. (2003). Lanzante (1996) also bases his procedure on the Wilcoxon statistic, and Yonetani and McCabe (1994) use the Lepage modificat”
Confidence sets for split points in decision trees (2007, background): “, in change-point analysis for climate data [12].”
Changepoint Detection in Periodic and Autocorrelated Time Series (2007, background): “For example, Lund and Reeves (2002) study a carbon dioxide series where a quadratic trend is apparent; the existence of a good reference series may render the inclusion of a trend component unnecessary.”
Homogenization of Temperature Series via Pairwise Comparisons (2009, methodology): “changes and trend changes (e.g., Vincent 1998; Lund and Reeves 2002; Wang 2003) are characterized by much lower powers of detection than the simpler dif-”
Detection of Undocumented Changepoints Using Multiple Test Statistics and Composite Reference Series. (2005, methodology): “If there is reasonable confidence that there is no trend in a time series, then estimation of the slope parameter can be eliminated and the two-phase model for {Yt} becomes (Lund and Reeves 2002)”
Attributes of Several Methods for Detecting Discontinuities in Mean Temperature Series (2006, methodology): “• Less apt (compared to SNHT and POTT) to identify breaks in series with a single imposed discontinuity • Even when modified as per Lund and Reeves (2002), the number of Type I errors was greater than expected • Offset magnitudes are more variable than those given by other procedures for single disc”