Jie Chen

Publications

3 Time Series Simulation by Conditional Generative Adversarial Net

Authors: Jie Chen, Rao Fu, Shutian Zeng, Yiping Zhuang, Agus Sudjianto

Abstract:

Generative Adversarial Net (GAN) has proved to be a powerful machine learning tool in image data analysis and generation. In this paper, we propose to use Conditional Generative Adversarial Net (CGAN) to learn and simulate time series data. The conditions include both categorical and continuous variables with different auxiliary information. Our simulation studies show that CGAN has the capability to learn different types of normal and heavy-tailed distributions, as well as dependent structures of different time series. It also has the capability to generate conditional predictive distributions consistent with training data distributions. We also provide an in-depth discussion on the rationale behind GAN and the neural networks as hierarchical splines to establish a clear connection with existing statistical methods of distribution generation. In practice, CGAN has a wide range of applications in market risk and counterparty risk analysis: it can be applied to learn historical data and generate scenarios for the calculation of Value-at-Risk (VaR) and Expected Shortfall (ES), and it can also predict the movement of the market risk factors. We present a real data analysis including a backtesting to demonstrate that CGAN can outperform Historical Simulation (HS), a popular method in market risk analysis to calculate VaR. CGAN can also be applied in economic time series modeling and forecasting. In this regard, we have included an example of hypothetical shock analysis for economic models and the generation of potential CCAR scenarios by CGAN at the end of the paper.

Keywords: Neural Network, Time series, conditional generative adversarial net, market and credit risk management

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 137
2 Optimal Location of the I/O Point in the Parking System

Authors: Jing Zhang, Jie Chen

Abstract:

In this paper, we deal with the optimal I/O point location in an automated parking system. In this system, the S/R machine (storage and retrieve machine) travels independently in vertical and horizontal directions. Based on the characteristics of the parking system and the basic principle of AS/RS system (Automated Storage and Retrieval System), we obtain the continuous model in units of time. For the single command cycle using the randomized storage policy, we calculate the probability density function for the system travel time and thus we develop the travel time model. And we confirm that the travel time model shows a good performance by comparing with discrete case. Finally in this part, we establish the optimal model by minimizing the expected travel time model and it is shown that the optimal location of the I/O point is located at the middle of the left-hand above corner.

Keywords: Response Time, optimal location, parking system, S/R machine

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 247
1 Travel Time Model for Cylinder Type Parking System

Authors: Jing Zhang, Jie Chen

Abstract:

In this paper, we mainly analyze an automated parking system where the storage and retrieval requests are performed by a tower crane. In this parking system, the S/R crane which is located at the middle of the bottom of the cylinder parking area can rotate in both clockwise and counterclockwise and three kinds of movements can be done simultaneously. We develop some mathematical travel time models for the single command cycle under the random storage assignment using the characteristics of this system. Finally, we compare these travel models with discrete case and it is shown that these travel models display a good satisfactory performance.

Keywords: tower crane, parking system, travel time model

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 423

Abstracts

3 Time Series Simulation by Conditional Generative Adversarial Net

Authors: Jie Chen, Rao Fu, Shutian Zeng, Yiping Zhuang, Agus Sudjianto

Abstract:

Generative Adversarial Net (GAN) has proved to be a powerful machine learning tool in image data analysis and generation. In this paper, we propose to use Conditional Generative Adversarial Net (CGAN) to learn and simulate time series data. The conditions include both categorical and continuous variables with different auxiliary information. Our simulation studies show that CGAN has the capability to learn different types of normal and heavy-tailed distributions, as well as dependent structures of different time series. It also has the capability to generate conditional predictive distributions consistent with training data distributions. We also provide an in-depth discussion on the rationale behind GAN and the neural networks as hierarchical splines to establish a clear connection with existing statistical methods of distribution generation. In practice, CGAN has a wide range of applications in market risk and counterparty risk analysis: it can be applied to learn historical data and generate scenarios for the calculation of Value-at-Risk (VaR) and Expected Shortfall (ES), and it can also predict the movement of the market risk factors. We present a real data analysis including a backtesting to demonstrate that CGAN can outperform Historical Simulation (HS), a popular method in market risk analysis to calculate VaR. CGAN can also be applied in economic time series modeling and forecasting. In this regard, we have included an example of hypothetical shock analysis for economic models and the generation of potential CCAR scenarios by CGAN at the end of the paper.

Keywords: Neural Network, Time series, conditional generative adversarial net, market and credit risk management

Procedia PDF Downloads 1
2 Optimal Location of the I/O Point in the Parking System

Authors: Jing Zhang, Jie Chen

Abstract:

In this paper, we deal with the optimal I/O point location in an automated parking system. In this system, the S/R machine (storage and retrieve machine) travels independently in vertical and horizontal directions. Based on the characteristics of the parking system and the basic principle of AS/RS system (Automated Storage and Retrieval System), we obtain the continuous model in units of time. For the single command cycle using the randomized storage policy, we calculate the probability density function for the system travel time and thus we develop the travel time model. And we confirm that the travel time model shows a good performance by comparing with discrete case. Finally in this part, we establish the optimal model by minimizing the expected travel time model and it is shown that the optimal location of the I/O point is located at the middle of the left-hand above corner.

Keywords: Response Time, optimal location, parking system, S/R machine

Procedia PDF Downloads 122
1 Shikonin Reduces Endometriosis by Inhibiting RANTES Secretion and Mononuclear Macrophage Chemotaxis

Authors: Jie Chen, Dong-ping Yuan, Lin Gu, Jun Long, Ni Jie, Ying-Li Shi

Abstract:

Endometriosis is a common disease in women of reproductive age, whose classic characteristic is mononuclear cell infiltration into lesions. Shikonin is an anti-inflammatory phytocompound from Lithospermum erythrorhizon, whose potential therapeutic effects for the endometriosis remain unclear. The working hypothesis was that shikonin can inhibit the development of endometriosis by the inhibition of chemotactic effect. Shikonin significantly inhibited the growth of human endometrial tissue implanted into mice (P<0.05). No observable adverse effects were found. The mouse regulated upon activation normal T-cell expressed and secreted (mRANTES) level in peritoneal fluid of animal endometriosis model was higher than that in normal SCID mice (P<0.05), and decreased dramatically after shikonin treatment in a dose-dependent manner (P<0.05). Peritoneal fluid from NOD/SCID mice treated with shikonin inhibited monocytes chemotaxis, which could be abolished by mRANTES antibody. In vitro, shikonin significantly inhibited RANTES expression of U937 cells cultured alone or co-cultured with human methothelail cells and endometrial stromal cells, and inhibited RANTES-induced chemotaxis of U937 cells (P<0.05). The present results suggest that shikonin can inhibit the development of endometriosis by mechanisms that at least include the inhibition of RANTES expression and decreased migration of mononuclear cells to lesions. Shikonin may be a useful and safe new approach for treating endometriosis.

Keywords: Endometriosis, shikonin, RANTES chemotaxis

Procedia PDF Downloads 213