✪ Arrowstreet Capital is a research-led discretionary global asset manager for institutional clients around the world.✪ We continually reinvest in research and infrastructure in order to maintain sustainable sources of competitive advantage and alpha.✪ It’s our belief that with the right combination of investment insights and quantitative tools, we can increase the risk-adjusted returns on client portfolios even after accounting for the transaction costs of active management.

✪ Language: Python, C++, C#✪ Database: MongoDB, SQL, KDB✪ API: Restful Interface, dask scheduler, docopt✪ System: AWS✪ Unit test: pytest, pylint, pydocstring✪ Distributed Computing: dask, multiprocessing, multithreading✪ OOP: Python meta programming, encapsulation, inheritance✪ Python tricks: decorator, generator, name mangling, memory cache, functools, collections✪ Version Control: GIT, JIRA✪ Algorithm: dynamic programming, BFS, DFS, divide and conquer, recursion, backtracking, binary search, bit manipulation, topological sort, graph✪ Machine Learning: Scikit-Learn, TensorFlow, PyTorch

✪ Developed a new VaR model for M&A Deals incorporating the deal break risk✪ Calculated the risk neutral deal break probability and compared time series models to project the loss given deal break✪ Conducted back-testing and analysis about risk-neutral probability of and correlations among different deal breaks✪ Implemented the model in Python and SQL to deal with data and calculate the VaR for any portfolio with M&A positions

✪ Worked on the project Robust Stochastic Approximation✪ Conducted research in theory and implementations of the classical, robust and mirror descent stochastic approximation approaches to stochastic optimization and convex-concave stochastic saddle point problems✪ Proved that a modified stochastic approximation approach outperforms the sample average approximation method for certain stochastic optimization and saddle point problems and showed the numerical results✪ Studied recursive algorithms including Kalman filter, Robbins-Monro algorithm and Kiefer-Wolfowitz procedure

✪ Worked on the project External Pay Gap and Corporate Policy✪ Collected financial data from Capital IQ and other databases and sorted it with VBA

✪ Wrote evaluation reports for companies planning IPOs in the NEEQ (considered as Chinese version of NASDAQ)Ÿ✪ Used linear regression models with different variables and analysis of variances to achieve reasonable valuations✪Ÿ Compared regressions (linear regression, polynomials, Laguerre polynomials, piecewise linear regression, kernel regression, etc.) and time series models (AR, neural network, etc.) to project the return and volatility of target company stock price

✪ Worked on the project A Kernel Method for Pricing and Hedging American Path-Dependent Options✪ Utilized Bloomberg Terminal to collect financial data regarding American path-dependent options and found the factors influencing the price which were not considered in the Black-Scholes Model

✪ Worked on the research project Government Ownership and Bank Risk✪ Tutored the Master's students attending the course Financial Accounting and grading their homework

✪ Tutored students on how to apply problem solving skills and analytical skills to their daily studies in courses related to Finance, Mathematics and Statistics✪ Solicited feedback from students and communicated with professors to improve teaching methods