Piotr Fryzlewicz's Publications and Software  
Preprints (year in brackets refers to the latest version)  
A changepoint approach to estimating the proportion of false null hypotheses in multiple testing. A. Kostic and P. Fryzlewicz (2023). In submission.  Software (R package MTCP) 
Fast and optimal inference for change points in piecewise polynomials via differencing. S. GavioliAkilagun and P. Fryzlewicz (2023). In submission.  Software (R package ChangePointInference) and code for reproducing the numerical results 
Doubly inhomogeneous reinforcement learning. L. Hu, M. Li, C. Shi, Z. Wu and P. Fryzlewicz (2022). In submission.  
Testing stationarity and change point detection in reinforcement learning. M. Li, C. Shi, Z. Wu and P. Fryzlewicz (2022). In submission.  Software 
Robust Narrowest Significance Pursuit: Inference for multiple changepoints in the median. P. Fryzlewicz (2023). Under revision.  Software 
Multiscale autoregression on adaptively detected timescales. R. Baranowski, Y. Chen and P. Fryzlewicz (2022). With supplement. Under revision.  Software (R package amar) 
Book  
Changepoint Detection and Data Segmentation  forthcoming book by P. Fearnhead and P. Fryzlewicz. Chapters posted here by the permission of the publisher. Any comments welcome.
Chapter 3  Detecting A Single Changepoint 

In Journals  
Multiple change point detection under serial dependence: Wild contrast maximisation and gappy Schwarz algorithm. H. Cho and P. Fryzlewicz (2023). Journal of Time Series Analysis, to appear.  Software 
Automatic changepoint detection in time series via deep learning. J. Li, P. Fearnhead, P. Fryzlewicz and T. Wang (2023). Journal of the Royal Statistical Society Series B (with discussion), to appear.  Software 
Detecting linear trend changes in data sequences. H. Maeng and P. Fryzlewicz (2023). With supplement. Statistical Papers, to appear.  Software (R package trendsegmentR) 
Narrowest Significance Pursuit: inference for multiple changepoints in linear models. P. Fryzlewicz (2023). With supplement. Journal of the American Statistical Association, to appear.  Software: R package nsp; see also zipped R code and data for reproducing the results reported in the paper 
Detection of multiple structural breaks in large covariance matrices. Y. Li, D. Li and P. Fryzlewicz (2023). With supplement. Journal of Business and Economic Statistics, 41, 846861.  Software (R package BSCOV) 
Exploiting disagreement between highdimensional variable selectors for uncertainty visualization. C. Yuen and P. Fryzlewicz (2022). With supplement. Journal of Computational and Graphical Statistics, 31, 351359.  Software: R package CSUV and Shiny app csuv 
Crosscovariance isolate detect: a new changepoint method for estimating dynamic functional connectivity. A. Anastasiou, I. Cribben and P. Fryzlewicz (2022). Medical Image Analysis, 75, 102252.  Software (R package ccid) 
Detecting multiple generalized changepoints by isolating single ones. A. Anastasiou and P. Fryzlewicz (2022). With supplement. Metrika, 85, 141174.  Software: R package breakfast; also in the older package IDetect. 
Regularizing axisaligned ensembles via data rotations that favor simpler learners. R. Blaser and P. Fryzlewicz (2021). Statistics and Computing, 31, 15.  Software (R package random.rotation) 
Detecting possibly frequent changepoints: Wild Binary Segmentation 2 and steepestdrop model selection. P. Fryzlewicz (2020). Journal of the Korean Statistical Society (with discussion), 49, 10271070. Rejoinder, 49, 10991105.  Software: R package breakfast; extra features in alpha version on Github. 
RankingBased Variable Selection for highdimensional data. R. Baranowski, Y. Chen and P. Fryzlewicz (2020). With supplement. Statistica Sinica, 30, 14851516.  Software (R package rbvs) 
Detection of gammaray transients with wild binary segmentation. S. Antier, K. Barynova, P. Fryzlewicz, C. Lachaud and G. MarchalDuval (2020). Monthly Notices of the Royal Astronomical Society, 493, 44284441.  
Predictive, finitesample model choice for time series under stationarity and nonstationarity. T. Kley, P. Preuss and P. Fryzlewicz (2019). Electronic Journal of Statistics, 13, 37103774.  Software (R package forecastSNSTS) 
NOVELIST estimator of large correlation and covariance matrices and their inverses. N. Huang and P. Fryzlewicz (2019). Test, 28, 694727.  Software (R package novelist) 
Regularized forecasting via smoothrough partitioning of the regression coefficients. H. Maeng and P. Fryzlewicz (2019). Electronic Journal of Statistics, 13, 20932120.  Software (R package srp) 
NarrowestOverThreshold detection of multiple changepoints and changepointlike features. R. Baranowski, Y. Chen and P. Fryzlewicz (2019). With supplement. Journal of the Royal Statistical Society Series B, 81, 649672.  Software: R package breakfast; also in the older package not. 
Likelihood ratio Haar variance stabilization and normalization for Poisson and other nonGaussian noise removal. P. Fryzlewicz (2018). Statistica Sinica, 28 (special issue in memory of Prof. Peter Hall), 28852901.  
Tailgreedy bottomup data decompositions and fast multiple changepoint detection. P. Fryzlewicz (2018). With supplement. Annals of Statistics, 46, 33903421.  Software (R package breakfast) 
Simultaneous multiple changepoint and factor analysis for highdimensional time series. M. Barigozzi, H. Cho and P. Fryzlewicz (2018). With supplement. Journal of Econometrics, 206, 187225.  Software (R package factorcpt) 
Complexvalued wavelet lifting and applications. J. Hamilton, M. Nunes, M. Knight and P. Fryzlewicz (2018). Technometrics, 60, 4860.  Software 
Multiple changepoint detection for nonstationary time series using Wild Binary Segmentation. K. Korkas and P. Fryzlewicz (2017). Statistica Sinica, 27, 287311.  Software (R package wbsts) 
SHAH: SHapeAdaptive Haar wavelets for image processing. P. Fryzlewicz and C. Timmermans (2016). Journal of Computational and Graphical Statistics, 25, 879898.  Software 
Random rotation ensembles. R. Blaser and P. Fryzlewicz (2016). Journal of Machine Learning Research, 17, 126.  Software (R package random.rotation) 
Relative liquidity and future volatility. M. Valenzuela, I. Zer, P. Fryzlewicz and T. Rheinländer (2015). Journal of Financial Markets, 24, 2548.  
Multiple changepoint detection for highdimensional time series via Sparsified Binary Segmentation. H. Cho and P. Fryzlewicz (2015). Journal of the Royal Statistical Society Series B, 77, 475507.  Software (R package hdbinseg) 
Wild Binary Segmentation for multiple changepoint detection. P. Fryzlewicz (2014). Annals of Statistics, 42, 22432281.  Software: R package breakfast; also in the older package wbs. 
Multiplechangepoint detection for autoregressive conditional heteroscedastic processes. P. Fryzlewicz and S. Subba Rao (2014). Journal of the Royal Statistical Society Series B, 76, 903924.  Software 
Adaptive trend estimation in financial time series via multiscale changepointinduced basis recovery. A. L. Schroeder and P. Fryzlewicz (2013). Statistics and Its Interface, 6, 449461.  
Highdimensional volatility matrix estimation via wavelets and thresholding. P. Fryzlewicz (2013). Biometrika, 100, 921938.  
A reflection of history: fluctuations in Greek sovereign risk between 1914 and 1929. O. Christodoulaki, H. Cho and P. Fryzlewicz (2012). European Review of Economic History, 16, 550571.  
Highdimensional variable selection via tilting. H. Cho and P. Fryzlewicz (2012). Journal of the Royal Statistical Society Series B, 74, 593622.  Software (R package tilting) 
TimeThreshold Maps: using information from wavelet reconstructions with all threshold values simultaneously. P. Fryzlewicz (2012). Journal of the Korean Statistical Society (with discussion), 41, 145159. Rejoinder: TimeThreshold Maps: using information from wavelet reconstructions with all threshold values simultaneously. P. Fryzlewicz (2012). Journal of the Korean Statistical Society, 41, 173175.  
Multiscale and multilevel technique for consistent segmentation of nonstationary time series. H. Cho and P. Fryzlewicz (2012). Statistica Sinica, 22, 207229.  Software 
Multiscale interpretation of taut string estimation and its connection to Unbalanced Haar wavelets. H. Cho and P. Fryzlewicz (2011). Statistics and Computing, 21, 671681.  
Thickpen transformation for time series. P. Fryzlewicz and H.S. Oh (2011). Journal of the Royal Statistical Society Series B, 73, 499529.  
Mixing properties of ARCH and timevarying ARCH processes. P. Fryzlewicz and S. Subba Rao (2011). Bernoulli, 17, 320346.  
The Dantzig selector in Cox's proportional hazards model. A. Antoniadis, P. Fryzlewicz and F. Letue (2010). Scandinavian Journal of Statistics, 37, 531552. See also this corrigendum.  Software 
Estimating linear dependence between nonstationary time series using the locally stationary wavelet model. J. Sanderson, P. Fryzlewicz and M. Jones (2010). Biometrika, 97, 435446.  
Consistent classification of nonstationary time series using stochastic wavelet representations. P. Fryzlewicz and H. Ombao (2009). Journal of the American Statistical Association, 104, 299312.  
Datadriven waveletFisz methodology for nonparametric function estimation. P. Fryzlewicz (2008). Electronic Journal of Statistics, 2, 863896.  Software 
A waveletFisz approach to spectrum estimation. P. Fryzlewicz, G. Nason and R. von Sachs (2008). Journal of Time Series Analysis, 29, 868880.  
Normalized leastsquares estimation in timevarying ARCH models. P. Fryzlewicz, T. Sapatinas and S. Subba Rao (2008). Annals of Statistics, 36, 742786.  
Unbalanced Haar technique for nonparametric function estimation. P. Fryzlewicz (2007). Journal of the American Statistical Association, 102, 13181327.  Software (R package unbalhaar) 
Bivariate hard thresholding in wavelet function estimation. P. Fryzlewicz (2007). Statistica Sinica, 17, 14571481.  Software 
GOES8 Xray sensor variance stabilization using the multiscale datadriven HaarFisz transform. P. Fryzlewicz, V. Delouille and G. Nason (2007). Journal of the Royal Statistical Society Series C, 56, 99116.  
Variance stabilization and normalization for onecolor microarray data using a datadriven multiscale approach. E. Motakis, G. Nason, P. Fryzlewicz and G. Rutter (2006). Bioinformatics, 22, 25472553.  Software (R package DDHFm) 
A HaarFisz technique for locally stationary volatility estimation. P. Fryzlewicz, T. Sapatinas and S. Subba Rao (2006). Biometrika, 93, 687704.  
HaarFisz estimation of evolutionary wavelet spectra. P. Fryzlewicz and G. Nason (2006). Journal of the Royal Statistical Society Series B, 68, 611634.  Software 
Parametric modelling of thresholds across scales in wavelet regression. A. Antoniadis and P. Fryzlewicz (2006). Biometrika, 93, 465471.  Software 
Modelling and forecasting financial logreturns as locally stationary wavelet processes. P. Fryzlewicz (2005). Journal of Applied Statistics, 32, 503528.  Software 
A HaarFisz algorithm for Poisson intensity estimation. P. Fryzlewicz and G. Nason (2004). Journal of Computational and Graphical Statistics, 13, 621638.  Software 
Forecasting nonstationary time series by wavelet process modelling. P. Fryzlewicz, S. Van Bellegem and R. von Sachs (2003). Annals of the Institute of Statistical Mathematics, 55, 737764.  Software 
Other  
Multiscale network analysis through tailgreedy bottomup approximation, with applications in neuroscience. X. Kang, P. Fryzlewicz, C. Chu, M. Kramer and E. Kolaczyk (2018). Proceedings of 2017 51st Asilomar Conference on Signals, Systems and Computers, 29 October1 November 2017, DOI: 10.1109/ACSSC.2017.8335617.  Software 
Invited discussion of "Random projection ensemble classification" by Cannings and Samworth. R. Blaser and P. Fryzlewicz (2017). Journal of the Royal Statistical Society Series B, 79, 1007.  
Invited discussion of "Statistical modelling of citation exchange between statistics journals" by Varin, Cattelan and Firth. P. Fryzlewicz (2016). Journal of the Royal Statistical Society Series A, 179, 4951.  
Discussion of "Multiscale change point inference" by Frick, Munk and Sieling. P. Fryzlewicz (2014). Journal of the Royal Statistical Society Series B, 76, 547548.  
On multizoom autoregressive time series models. P. Fryzlewicz (2013). Oberwolfach Reports 48/2013, 2124.  
Invited discussion of "Large covariance estimation by thresholding principal orthogonal complements" by Fan, Liao and Mincheva. P. Fryzlewicz and N. Huang (2013). Journal of the Royal Statistical Society Series B, 75, 648650.  
Wavelet methods. P. Fryzlewicz (2010). Invited overview paper in Wiley Interdisciplinary Reviews: Computational Statistics, 2, 654667.  
On the thickpen transformation for time series. P. Fryzlewicz (2010). Oberwolfach Reports 05/2010, 2730.  
Multiscale breakpoint detection in piecewise stationary AR models. H. Cho and P. Fryzlewicz (2008). Proceedings of IASC 2008, Yokohama, Japan, 58 December 2008.  
Locally stationary wavelet coherence with application to neuroscience. J. Sanderson and P. Fryzlewicz (2007). Proceedings of the 56th Session of the International Statistical Institute, Lisbon, Portugal, 2229 August 2007.  
Locally stationary wavelet coherence with application to neuroscience. J. Sanderson and P. Fryzlewicz (2007). Proceedings of the LASR 2007 Workshop  Systems Biology and Statistical Bioinformatics, Leeds, UK, 46 July 2007.  
A datadriven HaarFisz transform for multiscale variance stabilization. P. Fryzlewicz and V. Delouille (2005). Proceedings of the 13th IEEE/SP Workshop on Statistical Signal Processing, 1720 July 2005.  
A waveletbased model for forecasting nonstationary processes. S. Van Bellegem, P. Fryzlewicz and R. von Sachs (2003). In GROUP 24: Physical and Mathematical Aspects of Symmetries, Eds. JP. Gazeau, R. Kerner, JP. Antoine, S. Metens, JY. Thibon; IOP Publishing, Bristol.  Figure 
Theses  
Wavelet Techniques for Time Series and Poisson Data. P. Fryzlewicz (2003). Ph.D. Thesis, Department of Mathematics, University of Bristol, UK.  Software 
The Application of Linear Programming to American Option Valuation in the JumpDiffusion Model. P. Fryzlewicz (2000). M. Sc. Dissertation, Department of Mathematics, Wroclaw University of Technology, Poland. 