检索结果-内蒙古大学图书馆

Identification and Estimation for Matrix Time Series CP-factor Models

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chang, Jinyuan Du, Yue Huang, Guanglin Yao, Qiwei Joint Laboratory of Data Science and Business Intelligence Southwestern University of Finance and Economics Chengdu China Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing China Department of Statistics The London School of Economics and Political Science London United Kingdom

We propose a new method for identifying and estimating the CP-factor models for matrix time series. Unlike the generalized eigenanalysis-based method of Chang et al. (2023) for which the convergence rates may suffer from small eigengaps as the asymptotic theory is based on some matrix perturbation analysis, the proposed new method enjoy faster convergence rates which are free from any eigengaps. It achieves this by turning the problem into a joint diagonalization of several matrices whose elements are determined by a basis of a linear system, and by choosing the basis carefully to avoid near co-linearity (see Proposition 5 and Section 4.3 below). Furthermore, unlike Chang et al. (2023) which requires the two factor loading matrices to be full-ranked, the new method can handle rank-deficient factor loading matrices. Illustration with both simulated and real matrix time series data shows the advantages of the proposed new method. Copyright © 2024, The Authors. All rights reserved.

关键词： Time series

A data Analytics Approach Based on VAR Model for Digital Trade 4

学校读者我要写书评

暂无评论

A Data Analytics Approach Based on VAR Model for Digital Tra...

4th International Conference on Computing and Big data, ICCBD 2021

作者： Hu, Guang Chen, Yuhao Department of Data Science School of Statistics and Information Shanghai University of International Business and Economics School of Computer Science Fudan University Shanghai Key Laboratory of Data Science Fudan University China School of Statistics and Information Shanghai University of International Business and Economics China

ISBN: (纸本)9781450387194

Nowadays, the rapid development of digital trade drives the rise of digital trade platforms where the data analytics is a key function. For the data analytics function for the digital trade platform, this paper presents a data analytics approach based on vector autoregressive VAR model for the analysis on the influence factors of China's steel comprehensive price. In this paper, four variables, including comprehensive steel price, iron ore price, RMB exchange rate against US dollar and rebar price during the 14 years from 2005 to 2019, are selected to establish VAR model, and the relationship between the four variables is analyzed and the importance is compared. By analyzing the influencing factors of the comprehensive price of steel, this paper provides suggestions for the analysis of the trend of the comprehensive price of steel, the steel trade, and the management of the influencing factors of the comprehensive price of steel. © 2021 ACM.

关键词： data Analytics

HDTSA: An R Package for High-Dimensional Time Series Analysis

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chang, Jinyuan He, Jing Lin, Chen Yao, Qiwei Joint Laboratory of Data Science and Business Intelligence Southwestern University of Finance and Economics Chengdu China Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing China Department of Statistics The London School of Economics and Political Science London United Kingdom

High-dimensional time series analysis has become increasingly important in fields such as finance, economics, and biology. The two primary tasks for high-dimensional time series analysis are modeling and statistical inference, which aim to capture the underlying dynamic structure and investigate valuable information in the data. This paper presents the HDTSA package for R, which provides a general framework for analyzing high-dimensional time series data. This package includes four dimension reduction methods for modeling: factor models, principal component analysis, CP-decomposition, and cointegration analysis. It also implements two recently proposed white noise test and martingale difference test in high-dimensional scenario for statistical inference. The methods provided in this package can help users to analyze high-dimensional time series data and make reliable predictions. To improve computational efficiency, the HDTSA package integrates C++ through the Rcpp package. We illustrate the functions of the HDTSA package using simulated examples and real-world applications from finance and economics. Copyright © 2024, The Authors. All rights reserved.

关键词： Principal component analysis

Exploring Freight Loading Management by Deep Learning: a Case Study in Home Furnishing Industry

学校读者我要写书评

暂无评论

Annals of data science 2022年第2期9卷 213-228页

作者： Deng, Wei Patil, Rajvardhan Liu, Fangyao Daji, Ergu Shi, Yong College of Big Data Statistics Guizhou University of Finance and Economics Guizhou550025 China College of Computer and Information Science Arkansas Tech University RussellvilleAR68182 United States College of Information Science and Technology University of Nebraska at Omaha OmahaNE68182 United States Research Center on Fictitious Economy and Data Science The Chinese Academy of Sciences Beijing100190 China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing100190 China School of Economics and Management University of Chinese Academy of Sciences Beijing100190 China Nebraska Furniture Mart OmahaNE68124 United States Division of Science Research and Technology Southwest Minzu University Chengdu610041 China

Long short-term memory (LSTM) networks as state-of-the-art Deep Learning models, have achieved remarkable results in time series forecasting. However, they are less commonly applied to the industry of logistics. This paper presents two novel LSTM networks to predict the freight loading of routing areas, and the design of a smart loading management system is introduced. While most existing works on LSTM utilize the power of its prediction very well, our study shows less accurate and instable results if only LSTM network is applied, that arise from the big variation in the dataset. Instead, two constraints are inspired in this paper. The first constraint adds a time factor node to strengthen the correlation between predicted final loading units and booked loading units as close to delivery date. And the second proposal extends to parallel training by groupwise constraint. Experiments with across 3-year records in a national wide home furnishing retail company show that constraint LSTM networks significantly improve both the accuracy and stability of the prediction. Besides, the design of a smart loading management system will be shown, in which LSTM model plays the core rule to predict loading capacity, meanwhile the rule-based system triggers different events based on loading prediction and truck size. © 2021, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Long short-term memory

Improved Machine Learning Approach for Shanghai City's Secondhand Housing Price Prediction

学校读者我要写书评

暂无评论

Improved Machine Learning Approach for Shanghai City's Secon...

Computer science, Computer Engineering, & Applied Computing (CSCE), Congress in

作者： Guang Hu Yue Tang Rui Yi School of Statistics and Information Shanghai University of International Business and Economics School of Computer Science Fudan University Shanghai Key Laboratory of Data Science Shanghai China

This paper presents an improved machine learning approach for prediction of second-hand housing prices in Shanghai. It firstly builds the random forest model and the XGboost model with Shanghai second-hand housing transaction data, and then analyses the challenges in the improvement of the models' prediction accuracy and generalization performance. In the light of that, it introduces the Lasso model for variable selection, and deletes three variables with insignificant regression coefficients, and then rebuilds the random forest and XGboost prediction models. The experimental results show that the prediction accuracy and the generalization performance of the models is well improved.

关键词：

On the Noise Robustness of In-Context Learning for Text Generation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Gao, Hongfu Zhang, Feipeng Jiang, Wenyu Shu, Jun Zheng, Feng Wei, Hongxin Department of Statistics and Data Science Southern University of Science and Technology China School of Economics and Finance Xi’an Jiaotong University China National Key Laboratory for Novel Software Technology Nanjing University China School of Mathematics and Statistics Xi’an Jiaotong University China Department of Computer Science and Engineering Southern University of Science and Technology China

Large language models (LLMs) have shown impressive performance on downstream tasks by in-context learning (ICL), which heavily relies on the quality of demonstrations selected from a large set of annotated examples. Recent works claim that in-context learning is robust to noisy demonstrations in text classification. In this work, we show that, on text generation tasks, noisy annotations significantly hurt the performance of in-context learning. To circumvent the issue, we propose a simple and effective approach called Local Perplexity Ranking (LPR), which replaces the "noisy" candidates with their nearest neighbors that are more likely to be clean. Our method is motivated by analyzing the perplexity deviation caused by noisy labels and decomposing perplexity into inherent perplexity and matching perplexity. Our key idea behind LPR is thus to decouple the matching perplexity by performing the ranking among the neighbors in semantic space. Our approach can prevent the selected demonstrations from including mismatched input-label pairs while preserving the effectiveness of the original selection methods. Extensive experiments demonstrate the effectiveness of LPR, improving the EM score by up to 18.75 on common benchmarks with noisy annotations. Our code is available at https://***/ml-stat-Sustech/Local-Perplexity-Ranking © 2024, CC0.

关键词： Semantics

Iterative Privileged Learning for Multi-view Classification 10

学校读者我要写书评

暂无评论

Iterative Privileged Learning for Multi-view Classification

10th International Conference on Information Technology and Quantitative Management, ITQM 2023

作者： Fu, Saiji Tian, Yingjie Tang, Jingjing School of Economics and Management Beijing University of Posts and Telecommunications Beijing100876 China School of Economics and Management University of Chinese Academy of Sciences Beijing100190 China Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences Beijing100190 China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing100190 China MOE Social Science Laboratory of Digital Economic Forecasts and Policy Simulation UCAS Beijing100190 China School of Business Administration Faculty of Business Administration Southwestern University of Finance and Economics Sichuan 611130 China

Multi-view data is closely associated with privileged information, presenting opportunities for advancements in current studies. Firstly, they often overlook the dynamic nature of privileged information;Secondly, the current state-of-the-art privileged support vector machine, dSVM+, is primarily designed for single-view learning;Thirdly, some methods only address two-view classification problems;Finally, the performance enhancements achieved by these approaches remain unsatisfactory. To this end, this paper proposes a multi-view privileged support vector machine called MVdSVM+. The iterative learning strategy is put forward to update the privileged information dynamically. To our best knowledge, this is the first attempt of dSVM+ in multi-view learning. Meanwhile, it is designed for multiple views. Experiments show the effectiveness of MVdSVM+ over the benchmark methods. © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://***/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the Tenth International Conference on Information Technology and Quantitative Management.

关键词： Classification (of information)

Parameter identifiability of a within-host SARS-CoV-2 epidemic model

学校读者我要写书评

暂无评论

Infectious Disease Modelling 2024年第3期9卷 975-994页

作者： Junyuan Yang Sijin Wu Xuezhi Li Xiaoyan Wang Xue-Song Zhang Lu Hou Complex Systems Research Center Shanxi UniversityTaiyuan030006China Shanxi Key Laboratory of Mathematical Techniques and Big Data Analysis on Disease Control and Prevention Shanxi UniversityTaiyuan030006China School of Mathematics and Science Henan Normal UniversityXinxiang453000China School of Information Shanxi University of Finance and EconomicsTaiyuan030006China Agriculture and Animal Husbandry Technology Promotion Center of Xingan League Xingan League137400China

Parameter identification involves the estimation of undisclosed parameters within a system based on observed data and mathematical *** this investigation,we employ DAISY to meticulously examine the structural identifiability of parameters of a within-host SARS-CoV-2 epidemic model,taking into account an array of observable ***,Monte Carlo simulations are performed to offer a comprehensive practical analysis of model ***,sensitivity analysis is employed to ascertain that decreasing the replication rate of the SARS-CoV-2 virus and curbing the infectious period are the most efficacious measures in alleviating the dissemination of COVID-19 amongst hosts.

关键词： Structural identifiability Practical identifiability Sensitivity analysis The basic reproduction number

Estimation and inference of high-dimensional partially linear regression models with latent factors

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Shi, Yanmei Hao, Meiling Tang, Yanlin Guo, Xu School of Statistics Beijing Normal University Beijing China School of Statistics University of International Business and Economics Beijing China Key Laboratory of Advanced Theory and Application in Statistics and Data Science–MOE School of Statistics East China Normal University Shanghai China

In this paper, we introduce a novel high-dimensional Factor-Adjusted sparse Partially Linear regression Model (FAPLM), to integrate the linear effects of high-dimensional latent factors with the nonparametric effects of low-dimensional covariates. The proposed FAPLM combines the interpretability of linear models, the flexibility of nonparametric models, with the ability to effectively capture the dependencies among high-dimensional covariates. We develop a penalized estimation approach for the model by leveraging B-spline approximations and factor analysis techniques. Theoretical results establish error bounds for the estimators, aligning with the minimax rates of standard Lasso problems. To assess the significance of the linear component, we introduce a factor-adjusted projection debiased procedure and employ the Gaussian multiplier bootstrap method to derive critical values. Theoretical guarantees are provided under regularity conditions. Comprehensive numerical experiments validate the finite-sample performance of the proposed method. Its successful application to a birth weight dataset, the motivating example for this study, highlights both its effectiveness and practical relevance. Copyright © 2025, The Authors. All rights reserved.

关键词： Factor analysis