The goal of this project is to build a fully functional e-commerce website platform using a monolithic architecture design.
This project implements a high-concurrency, multi-threaded server with outstanding performance and stability.
This project has developed a cross-platform (Linux and Windows) MySQL and Redis server using C++. The server implements user login, registration, modification, and update data functions.
A high-concurrency server framework written in C++ that supports cross-platform usage (Linux/Windows).
At present, the CTP high-frequency quantitative trading platform only involves the basic functions, which include reading account configuration information, CTP penetration regulatory testing, account
This project aims to reproduce a simple network protocol stack, including ARP protocol, IP protocol, ICMP protocol, UDP protocol and time server, TCP protocol, and HTTP server components.
This is a FAT32 file system developed on the Windows platform, it is implemented layer by layer from the bottom up: driver layer, partition and FAT32 parsing, file operations and buffer management.
Implementing the Linux 0.11 kernel on the x86 architecture, completing the development of several submodules: bootloader/kernel loader/interrupt and exception handling/process scheduling...
This project contains three experiments on reinforcement learning. For instance, it demonstrates how to apply the DQN to the Lunar Lander environment and how DDQN can obtain better performance.
The course, authored by Prof. Jerzy in NYU, applies the R programming language to momentum trading, statistical arbitrage (pairs trading), and other active portfolio management strategies. The course implements volatility and price forecasting models, asset pricing and factor models, and portfolio optimization. The course will apply machine learning techniques, such as backtesting (cross-validation) and parameter regularization (shrinkage).
This course, taught by Prof.Jerzy in NYU, applies the R programming language to momentum trading, statistical arbitrage (pairs trading), and other active portfolio management strategies. The course implements volatility and price forecasting models, asset pricing and factor models, and portfolio optimization. The course applies machine learning techniques, such as backtesting (cross-validation) and parameter regularization (shrinkage).
Projects just for fun, eg. Apply Lasso Technique to Forecast Stock Movement, Apply Random Forest Technique to Forecast Stock Movement
Our team won the competition hold by IAQF in 2018. We submitted a solution titled, Implementing Momentum Strategy with Options: Dynamic Scaling and Optimization.
Chose 498 S&P 500 component stocks and categorized them according to the spread between their estimated and actual EPS, then calculated abnormal average returns of each group and plotted their graph individually by establishing a user-friendly menu
Calculated the implied volatility under dynamic price of options based on Newton-Raphson method, utilized Monte Carlo techniques for path-dependent derivative securities pricing
This is a trading platform including data retrieving, data formatting, feature selecting, backtesting..
