Modeling trend in temperature volatility using generalized LASSO

In this paper, we present methodology for estimating trends in spatio-temporal volatility. We give two algorithms for computing our estimator which are tailored for dense, gridded observations over both space and time, though these can be easily extended to other structures (time-varying network flows, neuroimaging). We motivate our methodology by applying it to a massive climate dataset and discuss the implications for climate analysis.

[1]  Peng Zeng,et al.  The dual and degrees of freedom of linearly constrained generalized lasso , 2015, Comput. Stat. Data Anal..

[2]  Stephen P. Boyd,et al.  Network Inference via the Time-Varying Graphical Lasso , 2017, KDD.

[3]  Gaston H. Gonnet,et al.  On the LambertW function , 1996, Adv. Comput. Math..

[4]  J. Besag.  Spatial Interaction and the Statistical Analysis of Lattice Systems , 1974 .

[5]  R. Engle.  Dynamic Conditional Correlation , 2002 .

[6]  T. Bollerslev,et al.  Generalized autoregressive conditional heteroskedasticity , 1986 .

[7]  P. Jones,et al.  No increase in global temperature variability despite changing regional patterns , 2013, Nature.

[8]  E. Fischer,et al.  Robust spatially aggregated projections of climate extremes , 2013 .

[9]  J. Hansen,et al.  Perception of climate change , 2012, Proceedings of the National Academy of Sciences.

[10]  Stephen P. Boyd,et al.  1 Trend Filtering , 2009, SIAM Rev..

[11]  R. Tibshirani.  Adaptive piecewise polynomial estimation via trend filtering , 2013, 1304.2986.

[12]  P. Huybers,et al.  Frequent summer temperature extremes reflect changes in the mean, not the variance , 2013, Proceedings of the National Academy of Sciences.

[13]  Stephen P. Boyd,et al.  Proximal Algorithms , 2013, Found. Trends Optim..

[14]  Peng Zeng,et al.  Geometry and Degrees of Freedom of Linearly Constrained Generalized Lasso , 2017 .

[15]  Kevin E. Trenberth,et al.  Climate variability and relationships between top‐of‐atmosphere radiation and temperatures on Earth , 2015 .

[16]  G. Tselioudis,et al.  Changes in extratropical storm track cloudiness 1983–2008: observational support for a poleward shift , 2011, Climate Dynamics.

[17]  J. Screen,et al.  Arctic amplification decreases temperature variance in northern mid- to high-latitudes , 2014 .

[18]  R. Tibshirani,et al.  Degrees of freedom in lasso problems , 2011, 1111.0653.

[19]  Ryan J. Tibshirani,et al.  Efficient Implementations of the Generalized Lasso Dual Path Algorithm , 2014, ArXiv.

[20]  V. Savage,et al.  Increased temperature variation poses a greater risk to species than climate warming , 2014, Proceedings of the Royal Society B: Biological Sciences.

[21]  Alexander J. Smola,et al.  Trend Filtering on Graphs , 2014, J. Mach. Learn. Res..

[22]  R. Tibshirani,et al.  The solution path of the generalized lasso , 2010, 1005.1971.

[23]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[24]  N. Shephard,et al.  Multivariate stochastic variance models , 1994 .