Quantile feature selection over correlated multivariate time series data has always been a methodological challenge and is an open problem. In this paper, we propose a general probabilistic methodology for feature selection in joint quantile time series analysis, under the name of quantile feature selection time series (QFSTS) model. The QFSTS model is a general structural time series model, where each component yields an additive contribution to the time series modeling with direct interpretations. Its flexibility is compound in the sense that users can add/deduct components for each times series and each time series can have its own specific valued components of different sizes. Feature selection is conducted in the quantile regression component, where each time series has its own pool of contemporaneous external predictors allowing "nowcasting". Creative probabilistic methodology in extending feature selection to the quantile time series research area is developed by means of multivariate asymmetric Laplace distribution, ``spike-and-slab" prior setup, the Metropolis-Hastings algorithm, and the Bayesian model averaging technique, all implemented consistently in the Bayesian paradigm. Different from most machine learning algorithms, the QFSTS model requires small datasets to train, converges fast, and is executable on ordinary personal computers. Extensive examinations on simulated data and empirical data confirmed that the QFSTS model has superior performance in feature selection, parameter estimation, and forecast.
翻译:相对于相关多变时间序列数据, 量度特征选择相对于相关多变时间序列数据, 总是一个方法上的挑战, 是一个开放的问题。 在本文中, 我们提出一个通用的概率方法, 用于在共四分位时间序列模型( QFSTS) 名称下, 共四分位特征选择时间序列( QFSTS) 模型中进行特征选择。 QFSTS 模型是一个一般的结构时间序列模型, 每个组成部分都为直接解释的时间序列模型做出附加贡献。 其灵活性在于用户可以为每个时间序列添加/ 下降元件, 并且每个时间序列可以拥有自己不同大小的具体价值组成部分。 特性选择是在量级回归分析中进行, 每个时间序列都有自己同时的外部预测器库,允许“ 播种 ” 。 将特性选择扩展到四分位时间序列研究区的创造性概率模型模型模型是通过多种变异式的配对时间序列分布、 “ spike- slab” 、 “Metopoli-Hashing” 和“Bayesian comlistrueal conistry ” 方法, 在Bayes real Bas、 和“ deliveral sal sal sal sal 数据采集和“ deal ”中, 和“ dealmaildalmaild dal” 和“ dealmaxald dald dal” 数据、 不同机器学习” 数据序列中, 和“pral 数据序列中, 和“ dalvialdalvialmaxaldaldaldaldaldald dald dald dald dald daldaldaldaldaldaldals” 和“ ” 和“ 和“ ” 和“ ” ” 和“Baldaldaldaldaldaldaldalbalbald” 和“ 和“ ” 和“ ” ” 和“praldaldal ” 和“Bald” 和“Baldaldaldaldaldald’s” 和“Bass” 和“Bas” 和“B