Modeling discrete stock price changes using a mixture of Poisson distributions

Jayasekare, Rasitha R.; Gill, Ryan; Lee, Kiseop

doi:10.1016/j.jkss.2016.01.002

Cited by 4 publications

(6 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Experiments on simulated and real-world data have shown that our approach outperforms competing methods that also target multilevel clustering tasks. Our developed model is based on the exponential family assumption with data distribution and thereby applies naturally to other data types; e.g., a mixture of Poisson distributions [8]. Finally, there are several possible directions for extensions from our work.…”

Section: Discussionmentioning

confidence: 99%

“…In this section, we provide a detailed algorithm for achieving local solution to regularized composite transportation barycenter in objective function in Eq. (8). To facilitate the discussion, we will remind the formulation of that objective function.…”

Section: B Regularized Composite Transportation Barycentermentioning

confidence: 99%

“…Given the formulation of Algorithm 3, the following theorem demonstrates that this algorithm determines the local solution of objective function (8) Theorem 3. The Algorithm 3 monotonically decreases the objective function (8) of regularized composite transportation barycenter until local convergence.…”

Section: B2 Local Convergence Of Algorithmmentioning

confidence: 99%

See 2 more Smart Citations

Probabilistic Multilevel Clustering via Composite Transportation Distance

Ho,

Huynh,

Phung

et al. 2018

Preprint

View full text Add to dashboard Cite

We propose a novel probabilistic approach to multilevel clustering problems based on composite transportation distance, which is a variant of transportation distance where the underlying metric is Kullback-Leibler divergence. Our method involves solving a joint optimization problem over spaces of probability measures to simultaneously discover grouping structures within groups and among groups. By exploiting the connection of our method to the problem of finding composite transportation barycenters, we develop fast and efficient optimization algorithms even for potentially large-scale multilevel datasets. Finally, we present experimental results with both synthetic and real data to demonstrate the efficiency and scalability of the proposed approach.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: B Regularized Composite Transportation Barycentermentioning

confidence: 99%

See 1 more Smart Citation

Probabilistic Multilevel Clustering via Composite Transportation Distance

Ho,

Huynh,

Phung

et al. 2018

Preprint

View full text Add to dashboard Cite

show abstract

“…In the domain of high-frequency trading (HFT), the most common method for producing probabilistic forecasts is by the stochastic modelling of price movements, an extensively researched topic in quantitative finance. [4]- [7] These approaches are however not extensively data-driven and rely on fitting closed-form theory-driven stochastic models, leading to major drawbacks such as sensitivity to regime shifts, intractability and lack of generalisation power. Due to these drawbacks, trading strategies driven by these models rely heavily on risk management to limit potential losses.…”

Section: Related Workmentioning

confidence: 99%

“…1) Continuous Markov Birth-Death process [6] 2) Poisson Mixture GLM [7] Benchmark 1 was chosen because it is a well established approach in the literature for the stochastic modelling of highfrequency price movements. In order to produce a probabilistic forecast in a form suitable for direct comparison with our proposed models, we implemented a process which draws samples from the fitted stochastic models and puts it through the emulator previously implemented for data collection to obtain an empirical distribution of the price movements.…”

Section: Benchmark Modelsmentioning

confidence: 99%

Deep Probabilistic Modelling of Price Movements for High-Frequency Trading

Lim¹,

Gorse²

2020

Preprint

View full text Add to dashboard Cite

In this paper we propose a deep recurrent architecture for the probabilistic modelling of high-frequency market prices, important for the risk management of automated trading systems. Our proposed architecture incorporates probabilistic mixture models into deep recurrent neural networks. The resulting deep mixture models simultaneously address several practical challenges important in the development of automated highfrequency trading strategies that were previously neglected in the literature: 1) probabilistic forecasting of the price movements; 2) single objective prediction of both the direction and size of the price movements. We train our models on high-frequency Bitcoin market data and evaluate them against benchmark models obtained from the literature. We show that our model outperforms the benchmark models in both a metric-based test and in a simulated trading scenario.

show abstract

Regression Models for Double Discrete Distributions

Alzaidani

Aly²

2022

Bull. Malays. Math. Sci. Soc.

View full text Add to dashboard Cite

Modeling discrete stock price changes using a mixture of Poisson distributions

Cited by 4 publications

References 24 publications

Probabilistic Multilevel Clustering via Composite Transportation Distance

Probabilistic Multilevel Clustering via Composite Transportation Distance

Deep Probabilistic Modelling of Price Movements for High-Frequency Trading

Regression Models for Double Discrete Distributions

Contact Info

Product

Resources

About