English proficiency is an important skill in the world today. For non-native En-glish speakers, a... more English proficiency is an important skill in the world today. For non-native En-glish speakers, an automated system capable of reviewing practice essays would assist English learners without requiring involvement of native essay graders. This project explored the use of an LSTM-RNN machine model for identifying existence of grammatical errors in essays written by non-native learners of English. The relatively high accuracy obtained suggests that using LSTM-RNN models is a promising approach and recommends future opportunities for refinements.
We survey the evolution, current situation, and future trends for gathering data about customers ... more We survey the evolution, current situation, and future trends for gathering data about customers in stores.
Insider trading is an ongoing problem in the financial marketplace with significant negative effe... more Insider trading is an ongoing problem in the financial marketplace with significant negative effects on the economic well being of the relevant market. This research would investigate whether it’s possible to detect insider trad- ing using emerging and established data science techniques. Eventually, the study will transition to discovering effective practical methods for detecting insider trading. As the availability of high quality data is the limiting factor for any contemporary data science effort, data will be researched in terms of what is available as well as would would be necessary but is not available to enable specific techniques. Sampling will be conducted chronologically; the most recent data, of whatever scope becomes necessary, will be used in whole. Instrumentation will consist primarily of SEC filings and historic stock price data, all publicly available. We'll also look into acquiring communication and relational data such as messaging across platforms standard in the industry, social network data, and web data. Once acquired, data will be reviewed and correlated by comparing buying and selling patterns to market patterns, social connections, executive role within the relevant company, and communication trends.
We test portfolio asset allocation strategies. Our portfolio gives 90% weight to assigned to Stra... more We test portfolio asset allocation strategies. Our portfolio gives 90% weight to assigned to Strategic Asset Allocation and remaining 10% of capital goes to Tactical Asset Allocation. The strategy is targeted to clients looking to invest in global equity and bond assets. Portfolio not only captures global bonds and equities premia but also tries to enhance the returns with strategic tilt as well as tactical allocation.
Given that our portfolio is targeted to clients looking to invest in 60%/40% global equity/bond mix, we have shown that our approach of using strategic tilt and tactical allocation in standard 60%/40% equity/bond outperform the benchmark, and has higher share ratio compared to the benchmark portfolio.
In this project, we implement two trading strategies. One is a statistical arbitrage strategy wit... more In this project, we implement two trading strategies. One is a statistical arbitrage strategy with sector ETFs. The other one involves S&P 500 index additions and deletions. Our statistical arbitrage strategy with sector ETFs utilizes the multiple linear regression model and outperforms the S&P 500 index.
In the S&P 500 index additions and deletions strategy, we devise a set of trading strategies which seek probability based on event-driven arbitrage. The strategies exploit price movements from stock additions to the S&P 500 Index.
Finding a solution to the 2-dimensional Ising model has necessitated the effort of many physicist... more Finding a solution to the 2-dimensional Ising model has necessitated the effort of many physicists. In this expository paper, we consider the development of the analytic solution to the Ising Model and further evolution of the solution such as the applicability of Monte Carlo techniques.
English proficiency is an important skill in the world today. For non-native En-glish speakers, a... more English proficiency is an important skill in the world today. For non-native En-glish speakers, an automated system capable of reviewing practice essays would assist English learners without requiring involvement of native essay graders. This project explored the use of an LSTM-RNN machine model for identifying existence of grammatical errors in essays written by non-native learners of English. The relatively high accuracy obtained suggests that using LSTM-RNN models is a promising approach and recommends future opportunities for refinements.
We survey the evolution, current situation, and future trends for gathering data about customers ... more We survey the evolution, current situation, and future trends for gathering data about customers in stores.
Insider trading is an ongoing problem in the financial marketplace with significant negative effe... more Insider trading is an ongoing problem in the financial marketplace with significant negative effects on the economic well being of the relevant market. This research would investigate whether it’s possible to detect insider trad- ing using emerging and established data science techniques. Eventually, the study will transition to discovering effective practical methods for detecting insider trading. As the availability of high quality data is the limiting factor for any contemporary data science effort, data will be researched in terms of what is available as well as would would be necessary but is not available to enable specific techniques. Sampling will be conducted chronologically; the most recent data, of whatever scope becomes necessary, will be used in whole. Instrumentation will consist primarily of SEC filings and historic stock price data, all publicly available. We'll also look into acquiring communication and relational data such as messaging across platforms standard in the industry, social network data, and web data. Once acquired, data will be reviewed and correlated by comparing buying and selling patterns to market patterns, social connections, executive role within the relevant company, and communication trends.
We test portfolio asset allocation strategies. Our portfolio gives 90% weight to assigned to Stra... more We test portfolio asset allocation strategies. Our portfolio gives 90% weight to assigned to Strategic Asset Allocation and remaining 10% of capital goes to Tactical Asset Allocation. The strategy is targeted to clients looking to invest in global equity and bond assets. Portfolio not only captures global bonds and equities premia but also tries to enhance the returns with strategic tilt as well as tactical allocation.
Given that our portfolio is targeted to clients looking to invest in 60%/40% global equity/bond mix, we have shown that our approach of using strategic tilt and tactical allocation in standard 60%/40% equity/bond outperform the benchmark, and has higher share ratio compared to the benchmark portfolio.
In this project, we implement two trading strategies. One is a statistical arbitrage strategy wit... more In this project, we implement two trading strategies. One is a statistical arbitrage strategy with sector ETFs. The other one involves S&P 500 index additions and deletions. Our statistical arbitrage strategy with sector ETFs utilizes the multiple linear regression model and outperforms the S&P 500 index.
In the S&P 500 index additions and deletions strategy, we devise a set of trading strategies which seek probability based on event-driven arbitrage. The strategies exploit price movements from stock additions to the S&P 500 Index.
Finding a solution to the 2-dimensional Ising model has necessitated the effort of many physicist... more Finding a solution to the 2-dimensional Ising model has necessitated the effort of many physicists. In this expository paper, we consider the development of the analytic solution to the Ising Model and further evolution of the solution such as the applicability of Monte Carlo techniques.
Uploads
Papers by Nina Kuklisova
Given that our portfolio is targeted to clients looking to invest in 60%/40% global equity/bond mix, we have shown that our approach of using strategic tilt and tactical allocation in standard 60%/40% equity/bond outperform the benchmark, and has higher share ratio compared to the benchmark portfolio.
In the S&P 500 index additions and deletions strategy, we devise a set of trading strategies which seek probability based on event-driven arbitrage. The strategies exploit price movements from stock additions to the S&P 500 Index.
Given that our portfolio is targeted to clients looking to invest in 60%/40% global equity/bond mix, we have shown that our approach of using strategic tilt and tactical allocation in standard 60%/40% equity/bond outperform the benchmark, and has higher share ratio compared to the benchmark portfolio.
In the S&P 500 index additions and deletions strategy, we devise a set of trading strategies which seek probability based on event-driven arbitrage. The strategies exploit price movements from stock additions to the S&P 500 Index.