Data Mining - Concepts and Techniques.

One of the key challenges in changepoint analysis is the ability to detect multiple changes within a given time series or sequence. The changepoint package has been developed to provide users with a choice of multiple changepoint search methods to use in conjunction with a given changepoint method and in particular provides an implementation of the recently proposed PELT algorithm. This article describes the search methods which are implemented in the package as well as some of the available test statistics whilst highlighting their application with simulated and practical examples. Particular emphasis is placed on the PELT algorithm and how results differ from the binary segmentation approach.

/pdf/changepoint-an-r-package-for-changepoint-analysis-4f9vs6okd3.pdf

changepoint: An R Package for Changepoint Analysis

The paper proposes Metropolis adjusted Langevin and Hamiltonian Monte Carlo sampling methods defined on the Riemann manifold to resolve the shortcomings of existing Monte Carlo algorithms when sampling from target densities that may be high dimensional and exhibit strong correlations. The methods provide fully automated adaptation mechanisms that circumvent the costly pilot runs that are required to tune proposal densities for Metropolis-Hastings or indeed Hamiltonian Monte Carlo and Metropolis adjusted Langevin algorithms. This allows for highly efficient sampling even in very high dimensions where different scalings may be required for the transient and stationary phases of the Markov chain. The methodology proposed exploits the Riemann geometry of the parameter space of statistical models and thus automatically adapts to the local structure when simulating paths across this manifold, providing highly efficient convergence and exploration of the target density. The performance of these Riemann manifold Monte Carlo methods is rigorously assessed by performing inference on logistic regression models, log-Gaussian Cox point processes, stochastic volatility models and Bayesian estimation of dynamic systems described by non-linear differential equations. Substantial improvements in the time-normalized effective sample size are reported when compared with alternative sampling approaches. MATLAB code that is available from http://www.ucl.ac.uk/statistics/research/rmhmc allows replication of all the results reported.

Riemann manifold Langevin and Hamiltonian Monte Carlo methods

The Bootstrap and Edgeworth Expansion

1. Introduction 2. Observation techniques 3. Description of ocean waves 4. Statistics 5. Linear wave theory (oceanic waters) 6. Waves in oceanic waters 7. Linear wave theory (coastal waters) 8. Waves in coastal waters 9. The SWAN wave model Appendices References Index.

/pdf/waves-in-oceanic-and-coastal-waters-6ofdu4icqi.pdf

Waves in Oceanic and Coastal Waters

Estimates of the wave directional spectrum of fetch-limited sea states are made from measurements made with a heave–pitch–roll buoy at the Maui location off the west coast of New Zealand. The fetch-limited sea states have significant wave heights between 0.5 and 4.5 m and are observed during persistent southeast wind events, which have a well-defined fetch of 200 km. The integrated properties of the estimated angular spreading distributions are in general agreement with those observed in previous studies. However, the angular distributions estimated for the Maui location are bimodal at frequencies greater than the spectral peak frequency. This result for deep water ocean waves is in contrast to the generally accepted unimodal angular distribution for wind seas, but it supports recently reported measurements of the angular distribution of fetch-limited waves in Lake George, Australia. Parametric relationships that describe the characteristics of the bimodal distributions are derived, and the impor...

Observations of the Directional Spectrum of Fetch-Limited Waves

https://www.lancaster.ac.uk/staff/jonathan/OcnEngExtRvw12.pdf

Statistical modelling of extreme ocean environments for marine design: A review

This paper concerns the collation, quality control, and analysis of single-point field measurements from fixed sensors mounted on offshore platforms. In total, the quality-controlled database contains 122 million individual waves, of which 3649 are rogue waves. Geographically, the majority of the field measurements were recorded in the North Sea, with supplementary data from the Gulf of Mexico, the South China Sea, and the North West shelf of Australia. The significant wave height ranged from 0.12 to 15.4 m, the peak period ranged from 1 to 24.7 s, the maximum crest height was 18.5 m, and the maximum recorded wave height was 25.5 m. This paper will describe the offshore installations, instrumentation, and the strict quality control procedure employed to ensure a reliable dataset. An examination of sea state parameters, environmental conditions, and local characteristics is performed to gain an insight into the behavior of rogue waves. Evidence is provided to demonstrate that rogue waves are not go...

Field Measurements of Rogue Water Waves

https://www.lancs.ac.uk/~jonathan/Kll09.pdf

Detection of changes in variance of oceanographic time-series using changepoint analysis

https://www.lancs.ac.uk/~jonathan/OcnEng10.pdf

Kevin Ewans

Papers

Observations of the Directional Spectrum of Fetch-Limited Waves

Statistical modelling of extreme ocean environments for marine design: A review

Field Measurements of Rogue Water Waves

Detection of changes in variance of oceanographic time-series using changepoint analysis

Joint modelling of wave spectral parameters for extreme sea states