Abstract The price discovery in finance markets has been the focus of many researches. On basis 1‐min high‐frequency returns, this paper sets out to examine dynamics between China Securities Index 300 index and its derivative products: futures, exchange‐traded fund (ETF), ETF options. Nonparametric nonlinear method thermal optimal path is adopted study. empirical evidence shows a clear difference ability concerning three products under investigation, their rank (sorted by strength) can be...

In hidden Markov chain (HMC) models, widely used for target tracking, the process noise and measurement are in general assumed to be independent Gaussian mathematical simplicity. However, independence assumptions do not always hold practice. For instance, a typical radar tracking application, is correlated over time as sampling frequency of generally much higher than bandwidth noise. addition, maneuvers outliers imply that non-Gaussian. To solve this problem, we resort triplet (TMC) models...

Abstract In this paper, we investigate whether the option features in convertible bonds lead to price discovery. On basis of minute‐level data from January 2019 December 2021, find that bond market contributes discovery stock by using thermal optimal path method, and are an important factor affecting ability. Regression analysis shows effect remains significant after controlling for a range variables, such as size bond, premium rate, information shock, volatility. addition, further explore...

Data loss is ubiquitous in practical engineering applications due to communication delay or congestion. rate a key metric evaluate the reliability of state estimation. To jointly estimate system and data rate, we propose class Gaussian-Beta filters for linear moderate nonlinear Gaussian state-space models with unknown probability measurement loss. In filters, arrival at each time formulated as binary random variable, which determined by classical threshold technology. addition, hidden are...

In practical system, the sensor biases may jump abruptly. An adaptive on-line algorithm is presented in this paper for situation. The can detect onset time and estimate level base on General Likelihood Ratio (GLR) test. Monte Carlo results show, our adaptively bias well estimation error will not increase remarkably as other previous registration algorithms. also converges to Cramer-Rao lower bound (CRLB) after jumping.

In this paper, a robust labeled multi-Bernoulli (RLMB) filter for the multi-target tracking (MTT) scenarios with inaccurate and time-varying process measurement noise covariances is proposed. The covariance are modeled as inverse Wishart (IW) distributions, respectively. state together predicted error inferred based on variational Bayesian (VB) inference. Moreover, closed-form implementation of proposed RLMB given linear Gaussian system predictive likelihood function calculated by minimizing...

Aiming at the problem of multi-target tracking under noise statistics mismatch, an adaptive δ-GLMB filter based on variational Bayesian (VB) approach is proposed. The joint distribution predicted state and corresponding error covariance matrix, measurement mean vector matrix are modeled as Normal-inverse Wishart (NIW) distributions, in which latent variables described Gamma distributions. In this paper, single-target filtering density expressed mixture Normal inverse (NNIWNIWGG), NNIWNIWGG...

Abstract In this paper, we analyze the role that trades and quotes play in price discovery. Based on tick‐level data for CSI 300 stock index futures, find contribution of to discovery does not differ from at low resolutions, but dominates high resolutions. This difference is influenced by spreads volume. Further analysis reveals intraday trending downward, up 31% first half‐hour. The adverse selection liquidity supply cost components significantly contribute dampen contribution, respectively.