  


l_1 Trend Filtering
SeungJean Kim (sjkimstanford.edu) Abstract: The problem of estimating underlying trends in time series data arises in a variety of disciplines. In this paper we propose a variation on HodrickPrescott (HP) filtering, a widely used method for trend estimation. The proposed l_1 trend filtering method substitutes a sum of absolute values (i.e., l_1norm) for the sum of squares used in HP filtering to penalize variations in the estimated trend. The l_1 trend filtering method produces trend estimates that are piecewise linear, and therefore is well suited to analyzing time series with an underlying piecewise linear trend. The kinks, or changes in slope, of the estimated trend can be interpreted as abrupt changes or events in the underlying dynamics of the time series. Using specialized interiorpoint methods, l_1 trend filtering can be carried out with not much more effort than HP filtering; in particular, the number of arithmetic operations required grows linearly with the number of data points. We describe the method, some of its basic properties, and give some illustrative examples. We show how the method is related to l_1 regularization based methods in sparse signal recovery and feature selection, and list some of extensions of the basic method. Keywords: detrending, l_1 regularization, HodrickPrescott filtering, piecewise linear fitting, sparse signal recovery, feature selection, time series analysis, trend estimation Category 1: Applications  Science and Engineering (Statistics ) Category 2: Convex and Nonsmooth Optimization (Other ) Category 3: Other Topics (Other ) Citation: Download: [PDF] Entry Submitted: 09/29/2007 Modify/Update this entry  
Visitors  Authors  More about us  Links  
Subscribe, Unsubscribe Digest Archive Search, Browse the Repository

Submit Update Policies 
Coordinator's Board Classification Scheme Credits Give us feedback 
Optimization Journals, Sites, Societies  