Questions tagged [change-point]
Methods that attempt to detect when a change occurs in a distribution, process, or function.
426
questions
0
votes
0
answers
17
views
Modified Structural Break Test
Consider a simple AR(1) model
$$y_t=c+\lambda y_{t-1}+\epsilon_t$$
For example using the so called Andrews supremum statistic, I can test for a single structural break in $\lambda$ at an unspecified ...
- 218
0
votes
0
answers
23
views
How to perform a breakpoint analysis on these two variables (Survey Score and Total # of Outages)?
Here is a snippet of the data:
I believe both of these are discrete variables. I want to first determine if a correlation exists. Would I utilize Kendall’s Rank Correlation? Second, I want to create ...
1
vote
0
answers
19
views
Estimating stationarity of evolving distribution
Consider a probability distribution $P_t$ at time $t \in \mathbb{N}$ which evolves in time, but is assumed to become stationary at some unknown time $t_0$, i.e $P_{t_0+1} = P_{t_0}$
To make matters ...
- 21
0
votes
1
answer
18
views
Chow F statistics using strucchange for ARMA(1,1)
I am looking to generate a series of Chow F statistics for an inflation rate series. I am using the strucchange package. Specifically, the Fstats() function. I want to calculate the Chow F statistics ...
- 11
1
vote
0
answers
15
views
Identifying concept drift in the presence of data drift
Usually concept drift is defined as a change in the conditional distribution of our target variable with our covariates,
$P_{0}(Y|X) \neq P_{t}(Y|X)$. It is also known as real concept drift or ...
- 1,035
0
votes
0
answers
19
views
Applying a rolling t.test on a single variable
I have a dataframe which looks like this:
...
0
votes
0
answers
16
views
How to compare trend for split datasets
I have product price data, at monthly frequency, available for a set of different products. Each of these series have a break point which is the point where we do some kind of intervention. The break ...
0
votes
0
answers
18
views
Time Series with Possible Discontinuity - Is the discontinuity significant?
I am just beginning to learn about statistics and statistical tests, so please forgive me if I'm using entirely the wrong terminology or I've missed something obvious.
I'm trying to solve a problem ...
- 1
1
vote
1
answer
49
views
Changepoints in linear regression (NOT piecewise regression)
I have two variables, X and Y, whose relationship can be described well by a linear regression. HOWEVER, this relationship changes every once in a while. It is not that the relationship changes ...
- 1,082
0
votes
0
answers
32
views
How to Deal with structural breakpoints?
well, recently, when estimating a regression model in R, I found multiple structural breaks. Reading on the subject, I noticed that such breaks can cause problems. Given this, what would be the ...
1
vote
0
answers
32
views
Prophet not fitting data correctly
I'm currently looking to build a time-series model to do change point detection on a churn dataset, in order to check that the covid19 measures taken during the pandemic had an influence on churn.
I'm ...
- 141
0
votes
1
answer
78
views
Breakpoints and Forecasting with R
I am new in R and Timeseries analysis and forecasting. I have 2 questions.
I am detecting three change points in my dataset.
...
- 1
0
votes
0
answers
22
views
How to implement change point in JAGS?
I know the package 'mcp' does Bayesian change point, but I want to use base JAGS. Apparently we can't have a random quantity specifying the density of the data. How does 'mcp' do it?
Here is my ...
- 428
2
votes
2
answers
450
views
Using PELT changepoint detection for observation counts data
I'm trying to detect changepoints in the number of observations (specifically the number of occurrences of x happening per day). Broadly speaking the events are independent and the time intervals ...
- 123
1
vote
0
answers
31
views
Using previous data in Bayesian change point detection
I'm interested in the detection of a single change point. I have phase I data, $X_{-n},\ldots, X_{-1}$, and I when subsequent observation arrive I want to find if their distribution has changed (...
- 1,035
1
vote
0
answers
35
views
Imposing a Constraint on Weights of Gaussian Linear Regression
I'm doing piece-wise Gaussian linear regression with one breakpoint in raw Python and Numpy and was wondering if it is possible to impose a condition on the weights. Here is the data and sample fit:
...
1
vote
1
answer
91
views
Change point detection
I have a specific question about the formulation of offline multiple change point detection given in Burg and Williams.
Where the change points are denoted $\{\tau_i\}$, and the slice of a time ...
- 223
0
votes
2
answers
79
views
Regression stats question
I am currently focusing on developing a way to predict the onset of sexual maturity in lake sturgeon by using the elemental composition in the fin ray.
Lake sturgeon spawn later in life (females ...
- 1
2
votes
1
answer
61
views
mgcv::gamm extract breakpoints
I want to model the relationship between individuals' weekly mean speed and the timing in the year (represented by the week number), to then extract the dates where we observe a change in the animals' ...
- 21
0
votes
0
answers
59
views
Cointegration with Multiple Structural Break in R
I am trying to find cointegrated stocks in S&P 500. I have data on all the 500 S&P stocks from 2000 to 2021 which includes multiple structural breaks.
I am doing the analysis in R.
Currently I ...
- 1
2
votes
3
answers
528
views
How to detect trend changes in noise time series?
wget -q -O- https://i.stack.imgur.com/xczQ1.gif | tail -c +43
R> with(f, plot(x, log10(y), type='l'))
I have a series whose ...
- 512
0
votes
0
answers
12
views
How should we deal with anomalies/structural breaks in the rear end of estimation period when forecasting outside the estimation window?
Suppose that we have successfully identified anomalies present near the end of our model estimation period. Also, suppose that the anomalies are linked to extreme events such as economic crises rather ...
- 11
0
votes
1
answer
49
views
Testing if certain points affect the time-series
I have a time series that describes the popularity of a product. I want to test if, when someone popular ( on social media ) publishes about the product ( positively or negatively ) it causes a "...
0
votes
0
answers
24
views
Quantitative method for determining ideal threshold to identify spikes (/anomalous events) in time-depth data?
This is building on a previous question asked here.
I have a load of time-depth data collected from some seabirds on a field trip. I am trying to classify diving behaviour by setting a fixed depth ...
- 73
0
votes
0
answers
23
views
Do structural breaks affect forecasting when using permutated data in an SVM?
1. I try to forecast the Industrial Production Index.
The data, both in the training set and the test set are randomly permuted and of course cross-validated (5-fold).
Under this setting, would ...
0
votes
0
answers
167
views
Bayesian update for a Gaussian distribution with unknown mean and variance
I am trying to implement the Bayesian Online Changepoint: https://arxiv.org/pdf/0710.3742.pdf in python. And I also have the step to update the sufficient statistics (variance and mean).
I have a ...
- 1
0
votes
0
answers
59
views
Help needed with ARDL Bounds Approach
I hope you are doing ok amidst this global pandemic.
I'm currently working on this research using the ARDL Bounds approach. My (to a degree self-imposed) "task" is to study the relationship ...
- 21
0
votes
0
answers
49
views
testing for stationarity with in the presence of structural breaks
I am in the search for structural breaks in a time series (price). I used a recent stata command that implements the Bai-Perron approach (command xtbreak), and found some breaks, which was expected.
...
- 3
0
votes
0
answers
59
views
Predictive posterior update for unknown mean and variance
I am trying to implement the Bayesian Online Changepoint: https://arxiv.org/pdf/0710.3742.pdf in python. And I also have the step to update the sufficient statistics (variance and mean).
The work is ...
- 1
0
votes
0
answers
48
views
How to model stock price time series using differential equations?
I work with stock price time series where I check for structural breaks in the series. To do that I fit simple models such as AR and ARIMA.
However, I was proposed to express the stock price in terms ...
- 219
0
votes
0
answers
11
views
What is the right way to handle a change in a time series data collection procedure?
We are building time series models to forecast freight volumes in the transportation industry.
So far, the data going into these models has been data on actual production levels.
At some point, we ...
- 125
0
votes
1
answer
23
views
Interpolating data during structural breaks?
What is the optimal strategy when you identify a structural break occuring very late in the series? I am using U.S. PCE data and unsurprisingly there is a significant break occuring at the outset of ...
3
votes
1
answer
58
views
What assumptions are required for structural breaks tests in time series?
Context
I'm doing research related to structural change in stock price time series. In order to test whether some chosen event is a structural break or the series has a structural change, I conduct a ...
- 219
1
vote
1
answer
226
views
How to handle a structural break of only one variable in a VAR model?
I am estimating a VAR-model with two variables: GDP real and Investments. Investments has a structural break, GDP real not.
As I understood it is only possible to add a dummy variable to both ...
- 65
2
votes
1
answer
309
views
How to interpret the Score-Based CUSUM test results?
Context
I'm doing an analysis of a price time series and checking for structural breaks (s.b. further on). One of these tests is the Score-Based CUSUM test. As far as I understand, this test is more ...
- 219
0
votes
1
answer
78
views
Sectional Linear trend for time series : Unknown breakpoints
I am looking for an algorithm which can detect trend with the following inputs
Minimum number of line segment.
Unknown breakpoints.
After searching , I found "Piece-wise linear fit" ...
- 11
0
votes
0
answers
31
views
Test for changing mean
I want to test whether a time series is better described as IID normal or as IID normal in two segments separated by a change point, with the mean and variance different in the two segments. If I have ...
- 586
0
votes
1
answer
54
views
How can I test two samples for different variance?
I am using an accelerometer to test for vibration. To clarify the physics, imagine an accelerometer (which measures acceleration) is sitting on top of a washing machine. On the spin cycle the ...
- 9
3
votes
0
answers
189
views
Finding changepoints in a GAM?
I'm using generalized additive modeling to investigate the relationship between two variables, X and Y. I want to find changepoints--i.e., X values at which the slope changes direction. I can get ...
0
votes
0
answers
68
views
Modelling a multivariate changepoint analysis
I'm trying to model a multivariate multiple change point analysis. The data I have is as follows:
Multiple sensors, positive count data, multiple change points and a time series.
For example, imagine ...
- 51
1
vote
1
answer
37
views
Structural break time series
Knowing that tests for structural breaks deal with breaks
in trend only, under what assumptions do you think it is reasonable to ignore
these breaks when modeling the ARMA process? Can they can ...
- 11
3
votes
2
answers
608
views
Time Series Chow test & Bai Perron test
I have a question about breakpoints test, that is to select which variables to put in the model ?
Example:
Y = C(constant) + Ut , Ut ~ N(0,1) ....first model
Y = C(constant) + Trend + Ut , Ut ~ N(0,...
- 69
2
votes
1
answer
150
views
Changepoint detection for normally distributed samples
I am looking for a test, or set thereof, that allows me to quantify if a time series has changed from one point in time to another.
Problem Description: I have a time series, consisting of ...
- 31
1
vote
1
answer
227
views
Segmented / piecewise / breakpoint regression derivation and output in R
I would like to do a segmented regression (or breakpoint or piecewise regression) with lines connecting at the breakpoint $c$. So
(a) $y = a_1 + b_1x$ for $x \leq c$
(b) $y = a_2 + b_2x$ for $x > ...
- 89