More Like this

1. A survey of numerical algorithms that can solve the Lasso problems

   https://doi.org/10.1002/wics.1602  

   Zhao, Yujie ; Huo, Xiaoming ( October 2022 , WIREs Computational Statistics)

   In statistics, the least absolute shrinkage and selection operator (Lasso) is a regression method that performs both variable selection and regularization. There is a lot of literature available, discussing the statistical properties of the regression coefficients estimated by the Lasso method. However, there lacks a comprehensive review discussing the algorithms to solve the optimization problem in Lasso. In this review, we summarize five representative algorithms to optimize the objective function in Lasso, including iterative shrinkage threshold algorithm (ISTA), fast iterative shrinkage‐thresholding algorithms (FISTA), coordinate gradient descent algorithm (CGDA), smooth L1 algorithm (SLA), and path following algorithm (PFA). Additionally, we also compare their convergence rate, as well as their potential strengths and weakness.

   This article is categorized under:

   Statistical Models > Linear Models

   Algorithms and Computational Methods > Numerical Methods

   Algorithms and Computational Methods > Computational Complexity

    

   more » « less
2. Sparse Signals in the Cross‐Section of Returns

   https://doi.org/10.1111/jofi.12733  

   CHINCO, ALEX ; CLARK‐JOSEPH, ADAM_D ; YE, MAO ( November 2018 , The Journal of Finance)

   This paper applies the Least Absolute Shrinkage and Selection Operator (LASSO) to make rolling one‐minute‐ahead return forecasts using the entire cross‐section of lagged returns as candidate predictors. The LASSO increases both out‐of‐sample fit and forecast‐implied Sharpe ratios. This out‐of‐sample success comes from identifying predictors that are unexpected, short‐lived, and sparse. Although the LASSO uses a statistical rule rather than economic intuition to identify predictors, the predictors it identifies are nevertheless associated with economically meaningful events: the LASSO tends to identify as predictors stocks with news about fundamentals.

    

   more » « less
3. Competing endogenous RNA network and prognostic nomograms for hepatocellular carcinoma patients who underwent R0 resection

   https://doi.org/10.1002/jcp.28634  

   Li, Yuntong ; Ma, Bingfen ; Yin, Zhenyu ; Liu, Pingguo ; Liu, Jianming ; Li, Jie ; Wang, Fuqiang ; Chen, Huimin ( April 2019 , Journal of Cellular Physiology)

   The prognosis of hepatocellular carcinoma (HCC) after R0 resection is unsatisfactory due to the high rate of recurrence. In this study, we investigated the recurrence‐related RNAs and the underlying mechanism. The long noncoding RNA (lncRNA), microRNA (miRNA), and messenger RNA (mRNA) expression data and clinical information of 247 patients who underwent R0 resection patients with HCC were obtained from The Cancer Genome Atlas. Comparing the 1‐year recurrence group (n = 56) with the nonrecurrence group (n = 60), we detected 34 differentially expressed lncRNAs (DElncRNAs), five DEmiRNAs, and 216 DEmRNAs. Of these, three DElncRNAs, hsa‐mir‐150‐5p, and 11 DEmRNAs were selected for constructing the competing endogenous RNA (ceRNA) network. Next, two nomogram models were constructed based separately on the lncRNAs and mRNAs that were further selected by Cox and least absolute shrinkage and selection operator regression analysis. The two nomogram models that showed a high prediction accuracy for disease‐free survival with the concordance indexes at 0.725 and 0.639. Further functional enrichment analysis of DEmRNAs showed that the mRNAs in the ceRNA network and nomogram models were associated with immune pathways. Hence, we constructed a hsa‐mir‐150‐5p‐centric ceRNA network and two effective nomogram prognostic models, and the related RNAs may be useful as potential biomarkers for predicting recurrence in patients with HCC.

    

   more » « less
4. A characterization method for low-frequency seismic noise in LIGO

   https://doi.org/10.1063/5.0122495  

   Valdes, G. ; Hines, A. ; Nelson, A. ; Zhang, Y. ; Guzman, F. ( December 2022 , Applied Physics Letters)

   We present a method to characterize the noise in ground-based gravitational-wave observatories such as the Laser Gravitational-Wave Observatory (LIGO). This method uses linear regression algorithms such as the least absolute shrinkage and selection operator to identify noise sources and analyzes the detector output vs noise witness sensors to quantify the coupling of such noise. Our method can be implemented with currently available resources at LIGO, which avoids extra coding or direct experimentation at the LIGO sites. We present two examples to validate and estimate the coupling of elevated ground motion at frequencies below 10 Hz with noise in the detector output.

    

   more » « less
5. Skillful statistical prediction of subseasonal temperature by training on dynamical model data

   https://doi.org/10.1017/eds.2023.2  

   Trenary, Laurie ; DelSole, Timothy ( January 2023 , Environmental Data Science)

   Abstract This paper derives statistical models for predicting wintertime subseasonal temperature over the western US. The statistical models are trained on two separate datasets, namely observations and dynamical model simulations, and are based on least absolute shrinkage and selection operator (lasso). Surprisingly, statistical models trained on dynamical model simulations can predict observations better than observation-trained models. One reason for this is that simulations involve orders of magnitude more data than observational datasets. 

   more » « less