1 / 20

By: David Johnston, James Mataras , Jesse Pirnat ,

By: David Johnston, James Mataras , Jesse Pirnat , Daniel Sanchez, Eric Shaw, Sean Vazquez, Brad Warren. Stevens Institute of Technology Department of Quantitative Finance: Professor Calhoun Department of Computer Science: Professor Klappholz. Table of Contents. Introduction Requirements

fabian
Download Presentation

By: David Johnston, James Mataras , Jesse Pirnat ,

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. By: David Johnston, James Mataras, Jesse Pirnat, Daniel Sanchez, Eric Shaw, Sean Vazquez, Brad Warren Stevens Institute of Technology Department of Quantitative Finance: Professor Calhoun Department of Computer Science: Professor Klappholz

  2. Table of Contents • Introduction • Requirements • Software Design • Features • Financial Models • Security • Challenges

  3. Introduction • System will enable more efficient and effective portfolios and risk management • Providing tools and analytics to drive investment decisions • Tools to support portfolio construction, position and trade analysis, risk metrics, and monitoring performance • System features a market model to help identify risks and trading opportunities • Client may leverage these tools to build a custom strategy based on quantitative analysis

  4. Objectives • Provide access to stock data without the need to pay for a subscription data source • Manage virtual trading accounts • Track portfolio performance • Risk forecasts and analytics • Analyze potential trades

  5. Functional Requirements • Portfolio Tracking • Input a portfolio and enter trades in the system • From the online feed for market data, system will provide updated quotes and charting capability for viewing portfolio performance • Portfolio will be made up of cash and long equities • Will not maintain a margin account • Risk Management • System shall provide portfolio risk metrics • Volatility forecasts • Scenario analysis • Sensitivities and correlations • Value-at-Risk • User may also drill down to position-level granularity

  6. Functional Requirements Cont. • Trade Analysis • With the risk management technology, the user shall be able to: • Analyze potential trades • Assess risk and return • See the effects on the portfolio as a whole

  7. Software Design

  8. Features and Usability • Multi platform usability • iPhone, Android, Tablet, Laptop/Desktop • Instant Access to Profile data and online data • Save all data on your profile online • Detailed graphing interface • No installation required!

  9. Factor Model • Stock returns are explained by a set of factors • As well as an idiosyncratic component • We use 4 factors • Total market return • Market cap • Value • Momentum • Makes for a tractable model • Dimensionality reduction • Intuition

  10. Regime Switching • Define a latent variable for the current regime • Returns in each regime are determined by a factor model with different parameters • Model dynamics using a Hidden Markov Model (HMM) • Regime transitions are described by a Markov process • Model calibration is data driven, using Machine Learning • Bayesian inference • Expectation-Maximization algorithm • Advantages over traditional factor models • Traditional factor models are Gaussian and stationary • Regime switching can generate behavior that better approximates empirical market dynamics • E.g. fat tails, heteroskedasticity, leverage effect, time-varying correlations

  11. Forecasts and Analytics • Monte Carlo simulation based on model • Forecasts the distribution of returns • Perform risk analysis based on simulation results • Return and volatility forecasts • Value at Risk • Marginal impact of individual positions and potential trades

  12. Security Requirements • Confidentiality • User Information • Account Password • Anonymity • Integrity • Data loss prevention • Data modification restrictions • Accessibility • In production: accessible anywhere at any time • Currently: only accessible on Stevens Campus

  13. Security Risks • User Accounts • Session Management • Authentication Mechanisms • Data Communication • Source Verification • HTTPS • Input Validation • XSS, JavaScript Injection, SQL Injection etc. • Denial of Service (DoS)

  14. Challenges: Implementation • Interfacing with live data • Integrating different technologies and algorithms in application • Wt C++ Framework • Caused significant setbacks • Was not familiar with any of the standard web tools and technologies: HTML5, CSS3, JavaScript, jQuery • Time constraints after moving off original Wt C++ web toolkit • Did not know about responsive web design • Gained familiarity and understanding of the traditional web technologies • Gained exposure to widely used Twitter Bootstrap framework and learned how responsive web design is done.

  15. Challenges: Financial Modeling • Model research and development • Literature research and applied quantitative analysis • Algorithm implementation • Efficiency is critical • Data acquisition • Bloomberg Terminal • Data cleaning • Size of historical dataset

More Related