MSR-CP: A Modified Segmented Regression Approach to Change Point Estimation in Rainfall Patterns in West Bengal
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Statistics. The day in statistics: methods, theory and the inference under them.
Tom: We're looking at the paper: "MSR-CP: A Modified Segmented Regression Approach to Change Point Estimation in Rainfall Patterns in West Bengal".
Jane: Here's the gist of the paper. MSR-CP: A Modified Segmented Regression Approach to Change Point Estimation in Rainfall Patterns in West Bengal presents a methodology for detecting change point in precipitation series using a smooth version of segmented regression model, which offers improved insights into evolving rainfall patterns and serves as a useful tool for climate data analysis and long-term forecasting A change point problem in time series analysis arises when the underlying data generating mechanism experiences a structural change at an unknown point in time The objective is to estimate the change point t0, along with the functional form of two regimes, i.e., f1(·) and f2(·), from the observed data For detecting mean changes, a common approach models E(Yt) as a piecewise linear regression function before and after t0 However, the unknown change point t0 enters the model as a nonlinear parameter making the estimation problem a nonlinear regression problem Existing Change Point Methods Existing methods include parametric approaches like Buishand Range (BR) Test and Standard Normal Homogeneity (SNH) Test, which rely on assumptions about the distribution of Yt Nonparametric approaches such as Pettitt’s Test and Sequential Mann-Kendall (SeqMK) test are also discussed These established methods are primarily designed to identify structural breaks characterized by abrupt shifts in the mean, but they are not well suited to detect gradual and systematic changes in trend common in climate data The Proposed MSR-CP Model To address the discontinuity issue in Threshold Regression Models (TRM), the segmented regression model (SRM) is used, but it suffers from non-differentiability at t0 which prohibits gradient-based optimization methods The Modified Segmented Regression for Change Point (MSR-CP) approximates the non-differentiable function with a smooth sigmoid function to retain continuity and differentiability with respect to t0 The model is formulated as Yt = β0 + β1t + β2(t − t0) · exp(δ(t − t0)) / 1 + exp(δ(t − t0)) + εt Estimation and Asymptotic Properties The estimation of parameters is performed by minimizing the sum of squares due to errors (SSE) with respect to β0, β1, β2 and x0 using the L-BFGS-B algorithm Under regularity conditions (A1)–(A3), Theorem 1 establishes the consistency of θˆn, meaning θˆn p→ θ∗ as n → ∞ Theorem 2 establishes asymptotic normality, stating that √n(θˆn − θ∗) d→ N4(0, σ2omega−1) as n → ∞ Simulation and Real Data Analysis Extensive simulation studies demonstrate that the modified segmented regression model for change point show consistent estimation of change points and model parameters, with decreasing bias and standard error as sample size increases The paper applies the proposed methods to 60 years of monsoon rainfall data from two districts of West Bengal, India, revealing a change point in the mid-1980s for both districts The results indicate that MSR-CP performs very close to SRM but with much lower computational time, and it is robust across different data generating scenarios including AR(1) errors The application of our methods to real-world precipitation data shows their practical relevance for climate research, enhancing the understanding of climate variability and improving long-term forecasting models The versatility and robustness of our model, demonstrated through simulations and real-data analysis, make it a valuable tool for examining various climate variables By detecting change points, especially in rainfall data, researchers can support water management, agricultural planning and adaptation strategies The study advances the field of climate change research by providing statistical methods for detecting change points in precipitation data, combining segmented regression with a smooth continuous model Acknowledgment The authors would like to thank West Bengal Pollution Control Board for providing access to the dataset used in this article The authors have nothing to report funding statement The authors declare no conflicts of interest References Alexandersson, H.
Tom: (1986).
Jane: A homogeneity test applied to precipitation data.
Tom: Journal of Climatology, 6 (6), 661–675.
Paper summary: Jane: So to wrap up this discussion on "MSR-CP: A Modified Segmented Regression Approach to Change Point Estimation in Rainfall Patterns in West Bengal," what does this paper actually mean for the field? It’s about taking a standard statistical tool and making it smarter to catch the slow changes that matter.
Tom: It’s about improving how we detect structural breaks in time series when those breaks aren't sudden jumps but gradual shifts, which is exactly what you see in climate data, and this paper gives us a methodology that is both statistically sound and computationally efficient enough for real-world analysis.
Lu: The authors extend the segmented regression model by incorporating a smooth and differentiable transition function based on Fong Huang et al., two thousand seventeen which addresses the non-differentiability issue inherent in traditional models, allowing for gradient-based optimization <ref:2611.07858#pg2>.
Meng: The practical implication is that researchers can now use this to better support water management or agricultural planning strategies because they’re getting more accurate change point detections for these long-term climate variables, and the computational efficiency is a big plus for deployment.
Lalam: The versatility and robustness of their model, demonstrated through simulations and real-data analysis, make it a valuable tool for examining various climate variables by combining segmented regression with a smooth continuous model.
Jane: It’s a versatile approach that moves beyond just abrupt shifts in mean to capture the slower trends that define climate change variability. It gives researchers new statistical methods for detecting those changes in precipitation data, which is useful for long-term forecasting models.
Tom: So yeah, this paper provides a statistical method that combines segmented regression with a smooth continuous model to detect change points, and its real-world application in West Bengal shows it’s a valuable tool for climate research.
Conclusion: Tom: So, we’ve been looking at this paper by MSR-CP, and it’s about detecting change points in rainfall data specifically for West Bengal. What does that title actually tell us?
Jane: It means they took a standard way to find shifts in data—segmented regression—and modified it to handle those gradual changes instead of just sudden jumps.
Tom: Right, and the authors are trying to use this smoothed version because the original segmented regression model has a problem where it gets jerky right at that change point, making it hard for computers to optimize.
Lu: The core idea is approximating that rough part with a smooth function, specifically a sigmoid shape, so the math works better for finding those parameters.
Meng: For me, the implication is practical; if we can get better estimates here than traditional tests like Buishand Range or Pettitt's Test, it could mean more reliable predictions for water management in that region.
Jane: Exactly. They did this on real monsoon rainfall data from West Bengal and found a change point around the mid-1980s for both districts <ref:2611.07858#pg2>. It’s not just theory; it’s applied to actual climate records from India.
Tom: And they showed that their modified approach, MSR-CP, performs very close to the standard segmented regression model but takes much less computational time. That's a big win for deployment, Meng would love that efficiency.
Meng: Yeah, if it runs faster and gives you similar results to something simpler, it’s definitely worth shipping into a production system. We need models that don't take forever to run on real traffic.
Lu: The paper confirms the consistency of their estimation method under various conditions—it works well even when the errors are not perfectly random, like those auto-regressive ones they tested with.
Jane: So, in simple terms, this paper gives us a statistical tool that’s more flexible for tracking slow climate trends than older methods we use now. It shows how to find those hidden shifts in rainfall patterns with a continuous model instead of just looking for sharp breaks.
Tom: It’s about providing a robust way to look at long-term climate variability, which is exactly what we need when building models that forecast things over the next few decades.
Jane: But the big question now is, how does this method compare statistically against those established tests like the Standard Normal Homogeneity Test?
Tom: That comparison is what we're going to dig into next, because understanding where it lands on a baseline tells us if this MSR-CP approach is actually an improvement for serious climate research.
Anirban Ghosh, Raju Maiti
Indian Statistical Institute
stat.ME, stat.AP
Submitted: 2026-11-05
Updated: 2026-11-05
Code: https://github.com/ganirban004/change-point-paper
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 58/100
The gist: MSR-CP: A Modified Segmented Regression Approach to Change Point Estimation in Rainfall Patterns in West Bengal presents a methodology for detecting change point in precipitation series using a…
Key concepts
- Change Point Problem
- This occurs when the underlying process generating time series data suddenly shifts at an unknown point in time. The goal is to find this exact point where the statistical behavior of the data changes, allowing researchers to model different regimes before and after that transition.
- Segmented Regression Model (SRM)
- This is a regression approach used to model time series data by fitting different functional forms for distinct periods. While useful, standard SRM models can have issues at the change point because the function becomes non-differentiable there, which complicates mathematical optimization.
- Modified Segmented Regression for Change Point (MSR-CP)
- This is a modification of SRM that uses a smooth sigmoid function to replace the sharp transition at the change point. This smoothness enables gradient-based optimization, making it easier and more stable to estimate parameters and accurately detect gradual shifts in climate data.
Terminology
Summary
MSR-CP: A Modified Segmented Regression Approach to Change Point Estimation in Rainfall Patterns in West Bengal presents a methodology for detecting change point in precipitation series using a smooth version of segmented regression model, which offers improved insights into evolving rainfall patterns and serves as a useful tool for climate data analysis and long-term forecasting
Introduction to the Problem
A change point problem in time series analysis arises when the underlying data generating mechanism experiences a structural change at an unknown point in time The objective is to estimate the change point t0, along with the functional form of two regimes, i.e., f1(·) and f2(·), from the observed data For detecting mean changes, a common approach models E(Yt) as a piecewise linear regression function before and after t0 However, the unknown change point t0 enters the model as a nonlinear parameter making the estimation problem a nonlinear regression problem
Existing Change Point Methods
Existing methods include parametric approaches like Buishand Range (BR) Test and Standard Normal Homogeneity (SNH) Test, which rely on assumptions about the distribution of Yt Nonparametric approaches such as Pettitt’s Test and Sequential Mann-Kendall (SeqMK) test are also discussed These established methods are primarily designed to identify structural breaks characterized by abrupt shifts in the mean, but they are not well suited to detect gradual and systematic changes in trend common in climate data
The Proposed MSR-CP Model
To address the discontinuity issue in Threshold Regression Models (TRM), the segmented regression model (SRM) is used, but it suffers from non-differentiability at t0 which prohibits gradient-based optimization methods The Modified Segmented Regression for Change Point (MSR-CP) approximates the non-differentiable function with a smooth sigmoid function to retain continuity and differentiability with respect to t0 The model is formulated as Yt = β0 + β1t + β2(t − t0) · exp(δ(t − t0)) / 1 + exp(δ(t − t0)) + εt
Estimation and Asymptotic Properties
The estimation of parameters is performed by minimizing the sum of squares due to errors (SSE) with respect to β0, β1, β2 and x0 using the L-BFGS-B algorithm Under regularity conditions (A1)–(A3), Theorem 1 establishes the consistency of θˆn, meaning θˆn p→ θ∗ as n → ∞ Theorem 2 establishes asymptotic normality, stating that √n(θˆn − θ∗) d→ N4(0, σ2omega−1) as n → ∞
Simulation and Real Data Analysis
Extensive simulation studies demonstrate that the modified segmented regression model for change point show consistent estimation of change points and model parameters, with decreasing bias and standard error as sample size increases The paper applies the proposed methods to 60 years of monsoon rainfall data from two districts of West Bengal, India, revealing a change point in the mid-1980s for both districts The results indicate that MSR-CP performs very close to SRM but with much lower computational time, and it is robust across different data generating scenarios including AR(1) errors
Conclusion
The application of our methods to real-world precipitation data shows their practical relevance for climate research, enhancing the understanding of climate variability and improving long-term forecasting models The versatility and robustness of our model, demonstrated through simulations and real-data analysis, make it a valuable tool for examining various climate variables By detecting change points, especially in rainfall data, researchers can support water management, agricultural planning and adaptation strategies The study advances the field of climate change research by providing statistical methods for detecting change points in precipitation data, combining segmented regression with a smooth continuous model
Acknowledgment
The authors would like to thank West Bengal Pollution Control Board for providing access to the dataset used in this article The authors have nothing to report funding statement The authors declare no conflicts of interest
References
Alexandersson, H. p. 6 For abrupt changes in the beginning and at the end of time series, a more appropriate
technique is the Standard Normal Homogeneity Test (SNHT). 2 Standard Normal Homogeneity (SNH) Test
Among parametric approaches, Standard Normal Homogeneity test (Alexandersson, 1986; Verstraeten et al. In this context, detecting change points in precipitation patterns using various statistical methods has drawn
significant attention in recent times. Change point detection in climate data has been carried out
using tests such as Buishand Range (BR), Standard Normal Homogeneity (SNH), Pettitt’s and the
Von Neumann Ratio (VNR), in addition to trend analysis based on the Mann-Kendall (MK) test
(Croitoru et al. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Areosa, W. D., McAleer, M., & Medeiros, M. p. 1 , 1988, Ter¨asvirta, 1994, 1996, Li et al. ,
2002, Saikkonen and Choi, 2004, Medeiros et al. , 1988, Ter¨asvirta, 1994, 1996, Li et al. ,
2002, Saikkonen and Choi, 2004, Medeiros et al. 1 Modified Segmented Regression for Change Point (MSR-CP)
In SRM (2), the term (t − t0)+ or equivalently the indicator function I(t > t0) introduces nondifferentiability at the change point t0, which prohibits the use of gradient-based optimization methods. To overcome the non-differentiability, our approach is to approximate the function (t − t0)+
with a smooth sigmoid function as
(t − t0)+ = (t − t0) · I(t > t0) ≈ (t − t0) ·
exp(δ(t − t0))
1 + exp(δ(t − t0)) (3)
where δ > 0 is a fixed tuning parameter controlling the sharpness of the transition.Where the paper says it
Where the paper says it
Where the paper says it
Alexandersson, H., A homogeneity test applied to precipitation data
Bouregaa, T. (2023). Change point detection and trend analysis of drought over algeria from 1901 to 2018. p. 2 , 2021 employed various parametric and nonparametric approaches for detecting change points
in precipitation and temperature data across Ethiopia, analyzing the data on annual, seasonal and
monthly scales. Bouregaa, 2023 further performed change point detection and trend analysis of
drought patterns in Algeria from 1901 to 2018.Where the paper says it
Britt, B. C. (2015). Stepwise segmented regression analysis: An iterative statistical algorithm to detect and quantify evolutionary and revolutionary transformations in longitudinal data. p. 0 MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data.Where the paper says it
Fong, Y., Di, C., Huang, Y. p. 2 Bouregaa, 2023 further performed change point detection and trend analysis of
drought patterns in Algeria from 1901 to 2018. In this paper, we study an extension of the segmented regression model that incorporates a
smooth and differentiable transition function based on Fong, Huang, et al. Bouregaa, 2023 further performed change point detection and trend analysis of
drought patterns in Algeria from 1901 to 2018. In this paper, we study an extension of the segmented regression model that incorporates a
smooth and differentiable transition function based on Fong, Huang, et al. , 2017) are a class of regression models in which the
relationship between the predictors and the outcome variable depends on certain threshold levels,
allowing the effect of predictors to change at specified time points, which are known as thresholdpoints or change-points. The most general TRM, stegmented regression model (StRM) (Britt, 2015; Fong, Huang, et al.Where the paper says it
Where the paper says it
Where the paper says it
Fong, Y., Huang, Y., Gilbert, P., & Permar, S. (2017). Chngpt: Threshold regression model estimation and inference. p. 8 Estimation of t0 can be done by minimizing the sum of squares due to errors
(SSE)
Xn
t=1
ε
2
t =
X
t0
t=1
(Yt − β0 − β1t)
2 +
Xn
t=t0+1
{Yt − (β0 + β2) − (β1 + β3)t}
2
with respect to β0, β1, β3, β4 and t0. A major limitation of the threshold regression model is that the sum of squared errors (SSE) is
discontinuous with respect to the change point parameter t0.Where the paper says it
Getahun, Y. S., Li, M.-H., & Pun, I.-F. (2021). Trend and change-point detection analyses of rainfall and temperature over the awash river basin of ethiopia. p. 2 , 2021 employed various parametric and nonparametric approaches for detecting change points
in precipitation and temperature data across Ethiopia, analyzing the data on annual, seasonal and
monthly scales. Bouregaa, 2023 further performed change point detection and trend analysis of
drought patterns in Algeria from 1901 to 2018.Where the paper says it
Hamilton, J., Time series analysis
Hillebrand, E., Medeiros, M. p. 1 , 1988, Ter¨asvirta, 1994, 1996, Li et al. ,
2002, Saikkonen and Choi, 2004, Medeiros et al. We propose an estimation procedure based on generalized least square (GLS). We also discuss the
asymptotic properties of the estimated parameters. To address these limitations, we discuss an extension of the traditional segmented
regression model enabling the detection of gradual shifts in trend. The developed methodology
is further extended to accommodate dependent error structures, which are common in time
series data.Where the paper says it
Where the paper says it
Where the paper says it
Jaiswal, R., Lohani, A., & Tiwari, H. (2015). Statistical analysis for change detection and trend assessment in climatological parameters. p. 1 In this context, detecting change points in precipitation patterns using various statistical methods has drawn
significant attention in recent times. Change point detection in climate data has been carried out
using tests such as Buishand Range (BR), Standard Normal Homogeneity (SNH), Pettitt’s and the
Von Neumann Ratio (VNR), in addition to trend analysis based on the Mann-Kendall (MK) test
(Croitoru et al.Where the paper says it
James, G., Witten, D., Hastie, T., & Tibshirani, R. (2014). An introduction to statistical learning: With applications in r. p. 0 Keywords: Change point; Precipitation data; Segmented regression; STR model. 1 Introduction
A change point problem in time series analysis arises when the underlying data generating mechanism
experiences a structural change at an unknown point in time. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github.Where the paper says it
Where the paper says it
Kendall, M., & Gibbons, J. p. 7 Then the test statistic is calculated as
U = max
t
|Ut
|,
and the p-value for a two-sided test is calculated as p = 2 exp( −6tˆ2
0
n(n+1) ), where tˆ0 is the estimated
change point, given by
tˆ0 = argmax
1≤t<n
Ut
3. 2 Sequential Mann-Kendall (SeqMK) Test
The Sequential Mann–Kendall test (Kendall and Gibbons, 1990; Schaber, 2005) is a nonparametric
method for detecting trend changes and identifying multiple possible change points without assuming
any distribution for the data. 1 Pettitt’s Test
Pettitt’s test (Pettitt, 1979; Verstraeten et al. , 2006) is a nonparametric rank-based method used to
detect a single point of shift in the location parameter of a time series.Where the paper says it
Where the paper says it
Khaliq, M. N., & Ouarda, T. p. 6 , n − 1, where
z1 =
1
t
X
t
i=1
Yi − Y¯
σ
, z2 =
1
n − t
Xn
i=t+1
Yi − Y¯
σ
. Then the test statistic to test the null hypothesis H0 : ∆ = 0 is given by
T = max
t
Tt
Critical values based on 1,000,000 Monte Carlo simulations are tabulated for T by Khaliq and Ouarda,
2007. The
estimated change point is given by
tˆ0 = argmax
1≤t≤n
|St
|
The Buishand range test is more sensitive for breaks in the middle of the series (Verstraeten
et al. For abrupt changes in the beginning and at the end of time series, a more appropriate
technique is the Standard Normal Homogeneity Test (SNHT).Where the paper says it
Where the paper says it
Kundu, S. K., & Mondal, T. p. 1 , 2015; Onyutha, 2016; Javari, 2016;
Suhaila and Yusop, 2018). Kundu and Mondal, 2019 conducted an analysis of long-term annual
2 The methodology
is subsequently applied to 60 years of monsoon rainfall data from two districts of West Bengal,
India. This approach offers improved insights into evolving rainfall patterns and serves as a
useful tool for climate data analysis and long-term forecasting. Extensive simulation studies are conducted to evaluate the performance and computational efficiency of the proposed approach relative to existing methods. The methodology
is subsequently applied to 60 years of monsoon rainfall data from two districts of West Bengal,
India.Where the paper says it
Where the paper says it
Where the paper says it
Lu, K.-P., & Chang, S. p. 1 , 2022 addressed the issue of change
point detection for count time series data by employing an INAR(1) process with time-varying covariates under STR setup. Lu and Chang, 2023 proposed a piecewise segmented regression model,
allowing for discontinuous jumps and connected change points in a single model. , f1(·) and f2(·), from the observed data. To find the change in means of
the distributions, a common approach is to model the mean, E(Yt), as a piecewise linear regression
function before and after the change point t0 as
E(Yt) = (
α1 + β1t for t ≤ t0
α2 + β2t for t > t0
.Where the paper says it
Where the paper says it
Luukkonen, R., Saikkonen, P. p. 1 , 1988, Ter¨asvirta, 1994, 1996, Li et al. ,
2002, Saikkonen and Choi, 2004, Medeiros et al. In the context of
change point analysis, the smooth transition regression (STR) models, initially proposed by Chan
and Tong, 1986, are further developed in Luukkonen et al. , 1988, Ter¨asvirta, 1994, 1996, Li et al. In the context of
change point analysis, the smooth transition regression (STR) models, initially proposed by Chan
and Tong, 1986, are further developed in Luukkonen et al. , 1988, Ter¨asvirta, 1994, 1996, Li et al. -Fari˜nas et al. , 2004 and Medeiros and Veiga, 2005 used smooth transition autoregressive (STAR) models into neural network setup.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Maiti, R., Li, J., Das, P., Liu, X., Feng, L., Hausenloy, D. J., & Chakraborty, B. (2023). A distributionfree smoothed combination method to improve discrimination accuracy in multi-category classification [PMID: 36384309]. Statistical Methods in Medical Research, 32 (2), 242–266. p. 0 MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data.Where the paper says it
Medeiros, M., Mendes, E., & Oxley, L. (Moment-based estimation of smooth transition regressions with endogenous variables [Moment Restriction-Based Econometric Methods]. p. 9 1 Modified Segmented Regression for Change Point (MSR-CP)
In SRM (2), the term (t − t0)+ or equivalently the indicator function I(t > t0) introduces nondifferentiability at the change point t0, which prohibits the use of gradient-based optimization methods. To overcome the non-differentiability, our approach is to approximate the function (t − t0)+
with a smooth sigmoid function as
(t − t0)+ = (t − t0) · I(t > t0) ≈ (t − t0) ·
exp(δ(t − t0))
1 + exp(δ(t − t0)) (3)
where δ > 0 is a fixed tuning parameter controlling the sharpness of the transition.Where the paper says it
Medeiros, M., & Veiga, A. p. 1 -Fari˜nas et al. , 2004 and Medeiros and Veiga, 2005 used smooth transition autoregressive (STAR) models into neural network setup. Bouregaa, 2023 further performed change point detection and trend analysis of
drought patterns in Algeria from 1901 to 2018. In this paper, we study an extension of the segmented regression model that incorporates a
smooth and differentiable transition function based on Fong, Huang, et al. -Fari˜nas et al. , 2004 and Medeiros and Veiga, 2005 used smooth transition autoregressive (STAR) models into neural network setup.Where the paper says it
Where the paper says it
Where the paper says it
Mira, S., & Escribano, A. p. 1 , 2008 and Areosa et al. Mira and Escribano,
2000 discussed the asymptotic properties of nonlinear least squares (NLS) under STR setup with
i. , 2008 and Areosa et al. Mira and Escribano,
2000 discussed the asymptotic properties of nonlinear least squares (NLS) under STR setup with
i.Where the paper says it
Where the paper says it
Nair, P., Chakraborty, A., Varikoden, H., Pavanathara, F., & Kuttippurath, J. (2018). The local and global climate forcings induced inhomogeneity of indian rainfall. p. 0 MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data.Where the paper says it
Nayak, S., Exploring the future rainfall characteristics over india from large ensemble global warming experiments. p. 1 Rice yield, in particular in West Bengal, heavily
depends on the timely and adequately distributed monsoon rainfall. Secondly, it helps in predicting
and managing water-related disasters like flooding and landslides, which are expected to worsen in
India due to climate warming (Nayak, 2023). MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data. 1 (R Core Team, 2026). The R codes are
available in author’s GitHub repository: https://github.Where the paper says it
Where the paper says it
Where the paper says it
Onyutha, C., Variability of seasonal and annual rainfall in the river nile riparian countries and possible linkages to ocean–atmosphere interactions. p. 2 and seasonal rainfall trends, along with change point detection of annual rainfall, using 102 years
(1901–2002) of monthly rainfall data from 18 stations across West Bengal. Similarly, Getahun et
al. MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data.Where the paper says it
Where the paper says it
Pettitt, A. N., A non-parametric approach to the change-point problem. p. 1 To find the change in means of
the distributions, a common approach is to model the mean, E(Yt), as a piecewise linear regression
function before and after the change point t0 as
E(Yt) = (
α1 + β1t for t ≤ t0
α2 + β2t for t > t0
. But the unknown change point t0 enters into the model as a nonlinear parameter making the estimation problem a nonlinear regression problem. MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data. Formally, a general formulation of the change point problem for a time series
{Yt
: t = 1, 2, . , n} is given by
Yt ∼
(
f1 for t ≤ t0
f2 for t > t0
∗Corresponding author: Anirban Ghosh, Applied Statistics Unit, Indian Statistical Institute, 203 B. 07858v1 [stat. ME] 5 Nov 2026Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Prais, S. J., & Winsten, C. B. (1954). Trend estimations and serial correlation. p. 0 We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data. To address these limitations, we discuss an extension of the traditional segmented
regression model enabling the detection of gradual shifts in trend.Where the paper says it
Priulla, A., & D’Angelo, N. p. 8 , 2017;
Fong, Huang, et al. , 2017; Liu and Qian, 2009; Priulla and D’Angelo, 2024) ensures continuity at t0
by enforcing constraints at the threshold point. Here only slope changes before and after t0. Thus, the segmented regression model improves upon the threshold model in two key aspects:
i it reduces the number of parameters and
9 MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data.Where the paper says it
Where the paper says it
Where the paper says it
R Core Team, R: A language and environment for statistical computing
Saikkonen, P., & Choi, I. p. 1 , 1988, Ter¨asvirta, 1994, 1996, Li et al. ,
2002, Saikkonen and Choi, 2004, Medeiros et al. , 2012; Chattopadhyay et al. In the context of
change point analysis, the smooth transition regression (STR) models, initially proposed by Chan
and Tong, 1986, are further developed in Luukkonen et al.Where the paper says it
Where the paper says it
Schaber, J., Phenology in germany in the 20th century: Methods, analyses and models [doctoralthesis]. p. 1 Understanding shifts in rainfall patterns is crucial for improving weather prediction models, optimizing water
management strategies and developing effective climate change adaptation measures. In this context, detecting change points in precipitation patterns using various statistical methods has drawn
significant attention in recent times.Where the paper says it
Seber, C., & Wild, C. (1989). Nonlinear regression. p. 1 To find the change in means of
the distributions, a common approach is to model the mean, E(Yt), as a piecewise linear regression
function before and after the change point t0 as
E(Yt) = (
α1 + β1t for t ≤ t0
α2 + β2t for t > t0
. But the unknown change point t0 enters into the model as a nonlinear parameter making the estimation problem a nonlinear regression problem.Where the paper says it
Sen, A., & Srivastava, M. p. 9 Consequently, derivative-free grid search based optimization methods are commonly employed. Here
Binary Segmentation (BS) (Scott and Knott, 1974; Sen and Srivastava, 1975) is employed. In this context, detecting change points in precipitation patterns using various statistical methods has drawn
significant attention in recent times. Change point detection in climate data has been carried out
using tests such as Buishand Range (BR), Standard Normal Homogeneity (SNH), Pettitt’s and the
Von Neumann Ratio (VNR), in addition to trend analysis based on the Mann-Kendall (MK) test
(Croitoru et al. MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data.Where the paper says it
Where the paper says it
Where the paper says it
Suhaila, J., & Yusop, Z. p. 1 , 2014; Jaiswal et al. , 2015; Onyutha, 2016; Javari, 2016;
Suhaila and Yusop, 2018). , 2021 employed various parametric and nonparametric approaches for detecting change points
in precipitation and temperature data across Ethiopia, analyzing the data on annual, seasonal and
monthly scales. Bouregaa, 2023 further performed change point detection and trend analysis of
drought patterns in Algeria from 1901 to 2018.Where the paper says it
Where the paper says it
Ter¨asvirta, T. (1994). Specification, estimation, and evaluation of smooth transition autoregressive models. p. 1 -Fari˜nas et al. , 2004 and Medeiros and Veiga, 2005 used smooth transition autoregressive (STAR) models into neural network setup. MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data.Where the paper says it
Where the paper says it
Ter¨asvirta, T. (1996). Modelling economic relationships with smooth transition regressions (SSE/EFI Working Paper Series in Economics and Finance No. p. 2 Bouregaa, 2023 further performed change point detection and trend analysis of
drought patterns in Algeria from 1901 to 2018. In this paper, we study an extension of the segmented regression model that incorporates a
smooth and differentiable transition function based on Fong, Huang, et al.Where the paper says it
Verstraeten, G., Poesen, J., Demar´ee, G., & Salles, C. (2006). Long-term (105 years) variability in rain erosivity as derived from 10-min rainfall depth data for ukkel (brussels, belgium): Implications for assessing soil erosion rates. p. 0 The methodology
is subsequently applied to 60 years of monsoon rainfall data from two districts of West Bengal,
India. This approach offers improved insights into evolving rainfall patterns and serves as a
useful tool for climate data analysis and long-term forecasting. MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data.Where the paper says it
Where the paper says it
Wikipedia contributors, Smooth maximum — Wikipedia, the free encyclopedia [[Online; accessed 30-October-2026]]
Yang, K., Zhang, G. p. 1 Nonlinear regression models have been widely used in practice for a variety of time series applications (Drignei et al. , 2008; Zhang et al. -Fari˜nas et al. , 2004 and Medeiros and Veiga, 2005 used smooth transition autoregressive (STAR) models into neural network setup. MSR-CP: A Modified Segmented Regression Approach to
Change Point Estimation in Rainfall Patterns in West Bengal
Anirban Ghosh∗1 and Raju Maiti2
1Applied Statistics Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
2Economic Research Unit, Indian Statistical Institute, 203 BT Road, Kolkata -
700108, India
Abstract
This study presents a methodology for detecting change point in precipitation series using a
smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in
climate data.Where the paper says it
Where the paper says it
Where the paper says it
Zarenistanak, M., Dhorde, A. G., & Kripalani, R. (2014). Trend and change point detection of annual and seasonal precipitation and temperature series over southwest iran. p. 2 , 2021 employed various parametric and nonparametric approaches for detecting change points
in precipitation and temperature data across Ethiopia, analyzing the data on annual, seasonal and
monthly scales. Bouregaa, 2023 further performed change point detection and trend analysis of
drought patterns in Algeria from 1901 to 2018.Where the paper says it
Zhang, Z., Albert, P. S., & Simons-Morton, B. (2012). Marginal analysis of longitudinal count data in long sequences: Methods and applications to a driving study. p. 0 The methodology
is subsequently applied to 60 years of monsoon rainfall data from two districts of West Bengal,
India. This approach offers improved insights into evolving rainfall patterns and serves as a
useful tool for climate data analysis and long-term forecasting. Formally, a general formulation of the change point problem for a time series
{Yt
: t = 1, 2, . , n} is given by
Yt ∼
(
f1 for t ≤ t0
f2 for t > t0
∗Corresponding author: Anirban Ghosh, Applied Statistics Unit, Indian Statistical Institute, 203 B.Where the paper says it
Where the paper says it
--- Page 38 ---
Table 6: Estimates, biases, standard errors and computation times (in milliseconds) for estimation of scaled change point when data generated from MSR-CP model with true parameters β = (1.5, 4, −8)⊤ with x0 = 0.5 and i.i.d. N(0, 2 2) errors
Table 7: Estimates, biases, standard errors and computation times (in milliseconds) for estimation of scaled change point when data generated from MSR-CP model with true parameters β = (−7, −5, 9)⊤ with x0 = 0.5 and AR(1) errors with ϕ = 0. p. 1 Hillebrand et al. , 2013 found the asymptotic
properties of quasi-maximum likelihood estimator (QMLE) considering time as a transition variable
when the errors are NIID.Where the paper says it
Table 8: Estimates, biases, standard errors and computation times (in milliseconds) for estimation of scaled change point when data generated from segmented regression model (SRM)
Table 9: Estimates, biases, standard errors and computation times (in milliseconds) for estimation of scaled change point when data generated from probit smoothing model
Table 10: Estimates and standard errors of model parameters along with 95% asymptotic confidence interval of the change point for MSR-CP with i.i.d. errors (for South 24 Parganas)
Table 11: Estimates and standard errors of model parameters along with 95% asymptotic confidence interval of the change point for MSR-CP with AR(1) errors (for South 24 Parganas)
Table 12: Estimated change point and computation time (in milliseconds) obtained through different methods for monsoon rainfall in South 24 Parganas
Table 13: Estimates and standard errors of model parameters along with 95% asymptotic confidence interval of the change point for MSR-CP with i.i.d. errors (for Coochbehar)
Table 14: Estimates and standard errors of model parameters along with 95% asymptotic confidence interval of the change point for MSR-CP with AR(1) errors (for Coochbehar)
Table 15: Estimated change point and computation time (in milliseconds) obtained through different methods for monsoon rainfall in Coochbehar
Figure 6
Figure 6: 95% confidence intervals for each methods when data are generated from true models
Figure 7
Figure 7: 95% confidence intervals for each methods when data are generated from alternative models
Figure 8
Table 10
Table 10: Estimates and standard errors of model parameters along with 95% asymptotic confidence interval of the change point for MSR-CP with i.i.d.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Implementation of a Modified Segmented Regression for Change Point (MSR-CP) model using a smooth sigmoid function allows AI systems to detect gradual shifts in trend in precipitation data that traditional methods miss, as it
enables the use of efficient gradient-based optimization techniques.
-
Extension of the MSR-CP model to accommodate AR(p) error structures through a feasible generalized least square approach enables AI systems to handle temporal dependence in time series data more accurately than i.i.d. assumptions allow, improving change point inference under real-world conditions where errors are not independent and identically distributed.
-
Development of an iterative estimation procedure for the MSR-CP model with AR(p) errors allows AI systems to simultaneously estimate the change point, model parameters, and autoregressive parameters by iteratively updating the error structure, leading to more robust parameter estimates compared to single-step methods.
-
Incorporation of asymptotic normality results provides AI systems with a mathematical framework for calculating confidence intervals for estimated change points, allowing them to report not just a single estimate but also a range of plausible values based on the sample size and model assumptions.
Abstract
This study presents a methodology for detecting change point in precipitation series using a smooth version of segmented regression model. We examine several existing change point detection methods and identify their limitations in capturing gradual changes commonly observed in climate data. To address these limitations, we discuss an extension of the traditional segmented regression model enabling the detection of gradual shifts in trend. The developed methodology is further extended to accommodate dependent error structures, which are common in time series data. Extensive simulation studies are conducted to evaluate the performance and computational efficiency of the proposed approach relative to existing methods. The methodology is subsequently applied to 60 years of monsoon rainfall data from two districts of West Bengal, India. This approach offers improved insights into evolving rainfall patterns and serves as a useful tool for climate data analysis and long-term forecasting.
Related papers
- Learning Robust Treatment Rules for Censored Data
- Neural Estimation of Summary Statistics for Simulation-Based Inference with Non-Asymptotic Guarantees
- Spectral divide-and-conquer MCMC for long stationary time series
- Improving the Efficiency of Subgroup Analysis in Randomized Controlled Trials with TMLE
- Personalized Federated Vector Autoregression with Personalization Diversity
- Doubly Robust and Efficient Calibration of Prediction Sets for Right-Censored Time-to-Event Outcomes