Methods that attempt to detect when a change occurs in a distribution, process, or function.
0
votes
0answers
10 views
Changepoint/Step Detection in Univariate Time Series
As a beginner to time series analysis, I'm trying to understand the best way of detecting the points at which my univariate time series shows a change in trend direction (see highlighted example).
I ...
0
votes
0answers
26 views
CUSUM algorithm and first derivative
I have a doubt about implementation of CUSUM algorithm. I found a Python implementation here.
It is based on a formulation of the CUSUM that basically tracks positive ($g^+$) as well as negative ($g^...
1
vote
1answer
63 views
1
vote
0answers
39 views
Hierarchical Version of Bayesian Change Detection Model in JAGS
I am trying to create a hierarchical changepoint detection model in JAGS, estimating group difference in changepoint based on individual changepoints in scores for an outcome variable (fictional in ...
0
votes
2answers
56 views
R changepoint, different number of outputs
I tried detect change points in R. I've ran following MWE in R.
...
0
votes
1answer
46 views
Finding statistically significant changes in a time series
What would be the proper way to determine statistically significant changes between time periods within a time series (between Yn and Yn+1)? I thought about taking the first difference and ...
1
vote
0answers
13 views
plotting the regression lines in each iteration (chngpt package) [closed]
I want to plot the regression lines for each iterations of the chngpt package. Link to chngpt package
Here's my sample data:
...
0
votes
0answers
105 views
Change detection point with python
How reliable is using KS test (https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.ks_2samp.html) for change-point detection (in a single vector)?
Is there a better way for this, in ...
1
vote
0answers
27 views
Time series clustering/segmentation based on pattern
I'm currently working with a database which contains several large PPG (pulse oximetry) and ECG time series. These series, however, contain segments within them which are highly contaminated by noise, ...
0
votes
0answers
24 views
Finding the likelihood of a normal distribution in a change point model
The simple change-point problem can be described as follows. Here it is assumed that both $p_1(y)$ and $p_2(y)$ are known completely.
$y_1,...,y_\tau|\tau $ are iid with distribution $p_1(y)$ and $y_{...
2
votes
0answers
88 views
Edge detection in time series
I have a time series (data here) which contains several square-wave jumps, as well as some physical signals of interest. An example is shown in the top panel of the figure below. There are square wave ...
0
votes
0answers
21 views
On the use of running total instead of raw data in change point detection
I am working with a percentage time series data recorded at irregular time step that generally shows or expected to show a decreasing trend with time (and at an increasing rate towards end of the time ...
0
votes
0answers
24 views
Understanding a cost function
I have been analyzing a cost function used in a point of change algorithm.
I am having trouble understanding the notion of a cost function and how it generates results for example.
...
0
votes
0answers
35 views
Find a “drop” in continuous time-series data
I have a dataset of a lot of time-series data of single observations. I am looking for a way to find a specific pattern in this data. As seen in the figure, the data does not change much over time, ...
0
votes
0answers
34 views
Hypothesis testing on dependent samples to identify change point
I want to perform hypothesis tests on the correlated observations. Practically speaking, suppose there is a sequence of data points $X_n$, which could be regarded as the observations from a certain ...
3
votes
0answers
32 views
How to estimate point of jump in a regression model?
So non-parametrically it is observed that once the temperature hits approximately 300K, there is a jump in conductivity of a material. Now the estimate is noisy and seems to shift slightly based on ...
1
vote
1answer
116 views
Finding changepoints in the movement of a car with respect to a lead car (using ecp package in R)
Main Question:
How do I apply e.divisive method from ecp package effectively on multivariate time-series, so that the resulting ...
3
votes
1answer
157 views
Need advice on change point (step) detection
I have a time series with lots of steps/jumps (data file here). A plot is given below. I would like to subtract an appropriate value for each of these square wave features to bring them back down to ...
4
votes
0answers
61 views
Detecting changes in large number of time-series that share seasonality
I have large number of time-series that are independent of each other, but share some seasonality patterns. I need to detect anomalies/changes (increased volume, change in mean), that appear in the ...
0
votes
1answer
47 views
sequences of different length
I have two sequences of coin tosses, which have different lengths. I want to compare their likelihoods under a model. But since the likelihood is a product of probabilities, the longer sequence ends ...
0
votes
1answer
30 views
Bayesian Information Criterion - Non-physical model selection
Following the work of Bai & Perron (1998) and others in detecting structural changes in time series, I am trying to select the breakpoints by using the Bayesian Information Criterion.
Basically, ...
0
votes
0answers
50 views
Time Series Analysis with Dependence between consecutive points
I have a time series data, let the number of available time series be 1000 and each time series consists of 1 million consecutive data points.
Sorry if I am not mathematically correct, but it seems ...
0
votes
2answers
45 views
Demonstrate change of predictor variable for a time series [duplicate]
I have the following problem:
Assume there are three time series $h(t)$, $g(t)$ and $f(t)$.
The hypothesis is that prior to $t_1$: $h(t)=a\cdot f(t-t_2)$
and after $t_1$: $h(t)=b \cdot g(t-t_3)$.
...
2
votes
0answers
49 views
Is it possible to use statistics to detect a rising temperature scenario?
It is required to monitor temperature data from a manufacturing environment. The particular use case is to detect if the temperature is rising beyond normal values to enable timely corrective action. ...
0
votes
1answer
330 views
Identify two major change points in time series data and summarise for several replicates in R
I'm working with temporal simulations (forward in time) and I would like to find the best way to detect the two major change points in this time series.
I have several replicates of the same ...
1
vote
1answer
93 views
How do I do a change point analysis on a sparse data set in python?
So, I have some data from video game playtests, where players were allowed to play a game at home for a week, and were asked to fill out a daily survey. In particular, they were asked to rate a ...
1
vote
1answer
47 views
Time series random chance detection
I have a time series data of monthly sales for the last 4 years. The sales for one month eg Feb 2017 is much higher same month last year or the last month sales of Jan 2017.
If i wanted to detect if ...
0
votes
0answers
57 views
How to statistically demonstrate a change of the predictor variable to a responding variable from time t on (time series)?
I have the following question:
I have three time series. Two form possible predictors (f(x) and g(x)) and a third is the responding function (h(x)). f(x) and g(x) exhibit different lags to h(x). Now I ...
1
vote
0answers
194 views
PyMC3 Hierarchical Change Point Model [closed]
Is it possible to run a hierarchical change point model in PyMC3? I keep getting a dimensions error. My approach may be naive:
...
4
votes
2answers
3k views
Time series anomaly detection
I am tasked to develop an anomaly detection system for data organised in many 1D (can be more than 1D if I choose, but I think that will complicate the problem even more) daily time series. The series ...
2
votes
4answers
888 views
Anchoring a linear regression to a specific data point in r
This is different, follow up question to someone else's question here:
I am very new to R and have no programming experience what so ever but I am holding my own for the data analysis we need in my ...
1
vote
0answers
82 views
Changepoint detection in a single unevenly spaced time series
I have a single time series, unevenly spaced. I would like to perform change point detection on it. Does it mean a non-parametric method is needed? Any pointers to which method would be valid? Also, ...
1
vote
0answers
132 views
How to detect distribution changes within population?
I have a time series data of populations, e.g. for each time point, I have the blood pressure of 1 million of rats. How can I monitor that the distribution of this population drift over time?
For ...
1
vote
1answer
154 views
When testing for structural changes, should a linear trend be identified from the data or the derivative of data?
There exists a linear trend in an otherwise non-linear set of time-series data. The linear trend can be identified using a change-point (or breakpoint for structural changes) analysis method. This ...
1
vote
0answers
106 views
detecting change point in the time series on a 2 dimensional space
I have a 2 dimensional geographic space. There are crime events occuring at different regions in the space over time. I am looking particularly at property crimes like burglary. If I look at the time ...
2
votes
2answers
111 views
How do I split a vector and minimize absolute error efficiently?
I have a vector $y\in\mathbb{R}^n$. I want to find a computationally efficient way to find a split point $k$, such that:
$\sum_1^k |y_i-\theta_1|+\sum_{k+1}^n |y_i-\theta_2|$ is minimised. The $\...
1
vote
3answers
275 views
Forecast time series with breakpoint using Holt-Winters (R)
I have a time series dataset for forecast using Holt-winters model, but my problem is the time series contains breaks or change point,
Should we remove the first break? Or is there any other solution ...
9
votes
1answer
205 views
How can I highlight noisy patches in a time series?
I have lots of time series data - water levels and velocities vs time. It is the output from a hydraulic model simulation. As part of the review process to confirm that the model is performing as ...
3
votes
0answers
284 views
What exactly is Beta in the Twitter Breakout Detection Algorithm?
I have been going through the R implementation of the EDM algorithm, Twitter Breakout Detection (here) and also the corresponding paper (here) and I'm not really understanding the concept of 'Beta' ...
4
votes
1answer
210 views
Judging flatness of time-series
I have multiple short (say, length <100 points) time-series as exemplified below. All the series are made of values measured in the same units. I need to find some criterion for judging their "...
1
vote
0answers
94 views
Forecasting in R and accounting for historical events (such as product price changes)
I'm trying to use the Holt-Winters model included with the R package 'forecast' to forecast a product's sales revenues, which includes seasonality.
In the past the product's price as changed, and ...
3
votes
1answer
2k views
simple trend analysis, time series
For purposes of a non-technical analysis of a simple time-series, what is a quick (but effective) technique / approach to understand (or even 'conclude' with some confidence level) that a trend has ...
1
vote
0answers
89 views
Transforming time series data before change point analysis in R
I have count data (non-financial) from 2010-2014 by week. I am interested in using R and changepoint package to find any significant points of time when the trend changed. I have two questions about ...
1
vote
1answer
176 views
Comparing the slopes of two regression lines where the two slopes are NOT independent (breakpoint)
so I am trying to see whether there is a breakpoint in my data set at a certain time. Basically I have 200 data points and am wondering if there is a breakpoint at t=100, ie whether the slopes of the ...
1
vote
0answers
80 views
Updating sufficient statistics parameter sets in Bayesian Inference (Changepoint Detection)
I am trying to implement and customize a changepoint detection method based on Bayesian Inference (referring to https://arxiv.org/pdf/0710.3742v1.pdf). Now I struggle understanding the conjugate prior ...
2
votes
1answer
1k views
What is change point analysis for At Most One Change(AMOC)?
Can any one explain me how can I choose the pen.value and penalty for the single change point detection i.e., for At Most One Change (AMOC). Is there any books that contains rich explanation of these ...
6
votes
1answer
559 views
Bayesian online changepoint detection (marginal predictive distribution)
I am reading the Bayesian online changepoint detection paper by Adams and MacKay (link).
The authors start by writing the marginal predictive distribution:
$$
P(x_{t+1} | \textbf{x}_{1:t}) = \sum_{...
3
votes
0answers
28 views
Tests for difference in predicted and actual trajectories in discontinuous longitudinal data
I am running a longitudinal study where the dose of a drug that interferes with cognitive performance is reduced across four days according to two differenct schedules. The first is a gradual ...
1
vote
2answers
189 views
Switchpoint calculation in linear models with probabilistic programming
My question is motivated by the switchpoint in texting frequency example at the end of the first chapter of "Probabilistic Programming and Bayesian Methods for Hackers". In that example, the rate of ...
4
votes
1answer
74 views
Identify changepoints in 1/0 sequence [duplicate]
Say I have a sequence that looks like this:
0 0 0 0 0 1 0 0 0 0 1 0 1 1 1 0 0 0 0 0 0 0 0 0 0
In general terms, I am interested in determining when there is an ...
