Correlated data are ubiquitous in today's data-driven society. While regression models for analyzing means and variances of responses interest relatively well developed, the development these correlations is largely confined to longitudinal data, a special form sequentially correlated data. This paper proposes new method analysis fully exploit use covariates general In renewed classroom highly unbalanced multilevel clustered with within-class within-school correlations, our reveals...

Abstract Expectiles have received increasing attention as a risk measure in management because of their coherency and elicitability at the level $\alpha\geq1/2$ . With view to practical assessments, this paper delves into worst-case expectile, where only partial information on underlying distribution is available there no closed-form representation. We explore asymptotic behavior expectile two specified ambiguity sets: one through Wasserstein distance from reference transforms problem convex...

We propose an autoregressive conditional Pareto (AcP) model based on the dynamic peaks over threshold method to a tail index in financial markets. Unlike score-based approach which is widely used many articles, we use exponential function process for its intuitiveness and interpretability. Probabilistic properties of AcP statistical parameter estimators maximum likelihood are studied this article. Real data show advantages AcP, especially, compared estimation volatility GARCH model, result...

In this paper, a second-order duality for non-differentiable minimax fractional programming is formulated by generalizing the one developed Husian et al. [Second order programming, Optim. Lett. 3 (2009), pp. 277–286], programming. The weak, strong and strict converse theorems are proved these programs under generalized η-bonvexity assumptions.

Early diagnosis significantly improves the survival rate in lung carcinoma patients. This study attempts to construct a predictive network between computational features and semantic of pulmonary nodules using online feature selection causal structure learning. In this paper, we exploit discovery based on streaming algorithm with symmetrical uncertainty algorithm. Different from traditional learning methods that usually obtain all advance then select optimal subset features, proposed...

Abstract Assessing conditional tail risk at very high or low levels is of great interest in numerous applications. Due to data sparsity tails, the widely used quantile regression method can suffer from variability especially for heavy-tailed distributions. As an alternative regression, expectile which relies on minimization asymmetric l 2 -norm and more sensitive magnitudes extreme losses than considered. In this article, we develop a new estimation by first estimating intermediate...

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

This paper proposes a novel censored autoregressive conditional Fréchet (CAcF) model with flexible evolution scheme for the time-varying parameters, which allows deciphering tail risk dynamics constrained by price limits from viewpoints of different preferences. The proposed can well accommodate many important empirical characteristics financial data, such as heavy-tailedness, volatility clustering, extreme event and limits. We then investigate via CAcF in price-limited stock markets, taking...