• Corpus ID: 251442825

Sensitivity of principal components to changes in the presence of non-stationarity

@inproceedings{Bette2022SensitivityOP,
  title={Sensitivity of principal components to changes in the presence of non-stationarity},
  author={Henrik M. Bette and Thomas Guhr},
  year={2022}
}
Non-stationarity affects the sensitivity of change detection in correlated systems described by sets of measurable variables. We study this by projecting onto different principal components. Non-stationarity is modeled as multiple normal states that exist in the system even before a change occurs. The studied changes occur in mean values, standard deviations or correlations of the variables. Monte Carlo simulations are performed to test the sensitivity for change detection with and without… 

Figures from this paper

Non-stationarity in correlation matrices for wind turbine SCADA-data and implications for failure detection

This work analyses high freqeuency SCADA-data from the Thanet offshore windpark in the UK and evaluates Pearson correlation matrices for a variety of observables with a moving time window, finding the main dependence of these states is shown to be on wind speed.

References

SHOWING 1-10 OF 40 REFERENCES

Which principal components are most sensitive in the change detection problem?

Principal component analysis (PCA) is often used in anomaly detection and statistical process control tasks. For bivariate normal data, we prove that the minor projection (the least varying

A systematic comparison of PCA-based statistical process monitoring methods for high-dimensional, time-dependent processes

These fundamental methods will be systematically compared on high-dimensional, time-dependent processes to provide practitioners with guidelines for appropriate monitoring strategies and a sense of how they can be expected to perform.

Uncovering the dynamics of correlation structures relative to the collective market motion

The measured correlations of financial time series in subsequent epochs change considerably as a function of time. When studying the whole correlation matrices, quasi-stationary patterns, referred to

Quasi-stationary states in temporal correlations for traffic systems: Cologne orbital motorway as an example

Traffic systems are complex systems that exhibit non-stationary characteristics. Therefore, the identification of temporary traffic states is significant. In contrast to the usual correlations of

Overview of PCA-Based Statistical Process-Monitoring Methods for Time-Dependent, High-Dimensional Data

A comprehensive review of the literature proposed to cope with high-dimensional and time-dependent features of statistical process monitoring, and a real-data example is presented to help the reader draw connections between the methods and the behavior they display.

Non-stationarity in correlation matrices for wind turbine SCADA-data and implications for failure detection

This work analyses high freqeuency SCADA-data from the Thanet offshore windpark in the UK and evaluates Pearson correlation matrices for a variety of observables with a moving time window, finding the main dependence of these states is shown to be on wind speed.

Model-Free Non-Stationarity Detection and Adaptation in Reinforcement Learning

This paper presents an adaptive RL algorithm able to detect changes in the environment or in the reward function and react to these changes by adapting to the new conditions of the task and tests its effectiveness in terms of non-stationarity detection and adaptation over a vanilla RL algorithm.

Taking climate change into account: Non‐stationarity in climate drivers of ecological response

Changes in the global climate system are creating increasingly non‐analogue climate conditions with expectations of non‐stationarity among climate drivers. Decoupling among climate drivers

Identifying States of a Financial Market

A definition of state is proposed for a financial market and it is found that a wide variety of characteristic correlation structure patterns exist in the observation time window, and that they can be classified into several typical “market states”.