Papers, Presentations, and Talks

17. Finding hidden structures using LDA

The slides of the talk I delivered at Indian Institute of Science on October 27, 2010 are available here. It is an introductory talk on Latent Dirichlet Allocation.

Assignment     Reference

16. Finding hidden structures using PLSA

The slides of the talk I delivered at Indian Institute of Science on October 22, 2010 are available here. It is an introductory talk on Probabilistic Latent Semantic Analysis.

15. Surveillance algorithms for detecting the lost object

Video surveillance systems are becoming a part of national infrastructure to prevent and control the alarmingly increasing terrorism. This report describes motion detection techniques and is applied for the detection of object lifting. [Download]

14.Vehicle Traffic Measurement using Computer Vision algorithms

The measurement of vehicle traffic is necessary for effective traffic control, monitoring and surveillance purposes. The video detection techniques use state of the art processing techniques to vehicle detection. Counting the number of vehicles is important for the design of adaptation of traffic signal for the congestion.This report describes counting algorithms based on computer vision techniques. [Download]

13. Design and implementation of decision strategy in agent-based market models

One of the key issues in agent-based modeling of stock market is how to incorporate decision strategy for various groups. Decision strategies using minority game approach have been designed and implemented. [Download]

12. Characterization in Agent-based market models

Agent-based market models are computational in nature to capture the features and dynamics of complex financial system. The characterization of agents in such models is a very important factor. In this report, we have described the modules for characterization of different type of agents in Agent-based market model. [Download]

11. Object tracking using radial basis function networks

The applications of visual tracking are broad in scope ranging from surveillance and monitoring to smart rooms. A robust object-tracking algorithm using Radial Basis Function (RBF) networks has been implemented using OpenCV libraries. The pixel-based color features are used to develop classifiers. The algorithm has been tested on various video samples under different conditions, and the results are analyzed. [Download]

10. Application of parallel processing in financial modeling

Financial models are becoming more and more complex in structures and their interactions. As several exotic financial products are introduced in the market, the need for high-performance computing environments and parallel machines has risen steadily over the period of years. In this report, we have outlined the various financial modeling paradigms and stressed the importance of use of parallel machines in implementing them. We have also described the steps involved in creating an integrated Market Simulator. [Download]

9. Doctoral thesis: System Dynamics Modeling of Stylized Features of Stock Markets 

The papers Stock markets - models and features, A stochastic model for individual's confidence bias, Models for limits to arbitrage and herding in stock markets, Emergent behavior of a system with minority game evolution rules, and A computational market model incorporating different types of traders go into my doctoral thesis as chapters. A brief summary of the thesis is posted in my blog. 

Download the thesis from here (pdf file).

8. Stock Market Modeling: A Synergetic Approach

This was one of my early papers in the area of stock markets. It explored the issue of methodological aspects of stock market modeling where individual actions at micro-level and effects at macro-level interact strongly. The paper also talked about the emergent structures that arise due to interactions at micro- and macro-levels. My doctoral thesis was a kind of exploration that took off from a line of thinking stated philosophically in this paper. There are few more ideas in this paper which are not explored till now. I will probably work on them in future.

Download the paper from here (pdf file).

7. A Computational Market Model Incorporating Different Types of Traders 

Computational models of financial markets focus on dynamic interactions among diverse set of traders. The motivation behind the proposed model is to use the same framework representing the trader either as a first-order or second-order system, and also to incorporate the minority game rules for the rule-based traders through switching dynamical system. The model generates the price process with the semblance of fat-tail property only if there is heterogeneity amongst the traders. Heterogeneity also reduces the volatility if the noise traders are dominated by other types of traders.

Download the paper from here (pdf file).

6. Emergent Behavior of a System with Minority Game Evolution Rules 

A switching dynamical system approach has been proposed with the minority game rules incorporated in. Analysis and simulations show that proposed model is flexible enough to consider complex behavior of markets. The emergent behavior of such system seems to capture the characteristics features of the minority game. The volatility variation can be interpreted from individual choices and actions.

Here is the link to download the paper (pdf file). 

5. Models for Limits to Arbitrage and Herding in Stock Markets 

A system dynamical model to capture the limits to arbitrage is proposed. The impact of the information on the arbitrage can be incorporated in the proposed model. The explicit condition for no arbitrage is determined using the trader's profile. The ways in which different delays can be implemented in the new framework have been described. We have attempted to model herding through an observer interpretation in the dynamical system framework. Herding is interpreted as manifestation of reduced order dynamics of a given system. The risk associated is related to the asymptotic behavior of the error signal. 

Download the paper from here (pdf file). 

4. A Stochastic Model for Individual's Confidence Bias

The confidence bias is the systematic overestimating or underestimating the accuracy of one's decisions and knowledge. A model is proposed where the confidence bias of an individual is assumed to be dependent upon the outcome of past decision. The impact of this confidence on the present capital is considered through the concept of virtual gain, where we have incorporated the aspiration level of Prospect Theory. The effect of confidence on future decision is brought out explicitly. 

Download the paper from here (pdf file).

3. Stalking a Stock Market: System, Signature, and Supposition 

This was the talk delivered at Panorama-2007 Research Symposium, Centre for Electronics Design and Technology, Indian Institute of Science, Bangalore on March 07, 2007. The duration of the talk was about an hour. In this talk, some general signatures of a stock market system were discussed. The system theoretic view of the stock market was stressed. The different ways to model those signatures were also elaborated. The limitation of expected utility concept from the point of view of Prospect theory was also discussed with an example. 

You may download the presentation slides of the talk from this link (ppt file). 

2. Stock Markets - Models and Features

Stock market is a complex system which is a part of the much broader economic system. Stock market and other financial markets, such as foreign exchange market, play an important role in the modern economy. The rewards, for investors, which the stock market holds out are enormous. It is also equally capable of bringing calamitous ruin. People have sought methods to understand the state and trend of the market. System theoretical framework is fairly a general one where different inter-relationships can be incorporated, and many complex behaviors of the system can be interpreted as due to the fundamental mechanisms of feedback and coupling among components. There is no single framework to model various aspects of the stock market. There are different approaches to model different characteristics of the market.

Here is the download link (pdf file). It is reported at Finance Research website that this paper was downloaded maximum number of times from their site for the last quarter (Oct-Dec) of 2007.

1. Stochastic Systems 

There were special lecture sessions on Control Systems at Centre for Electronics Design and Technology, Indian Institute of Science, Bangalore in May 2000. Dinesh Pai presented control system design techniques. Venkatesh delivered a talk on PID controller. I lectured on stochastic systems in three parts. The first talk dealt with how to include stochastic nature into a system, why Ito calculus is different from that of deterministic one, and various ways to evaluate stochastic integration. The second talk considered systems with stochastic inputs, spectral characterization, and iterated least squares. The third talk was on minimum variance control strategies with few simulation examples. The presentation slides (ppt files) of the three talks are here - Talk-1, Talk-2, and Talk-3.