Sign in

Data Project

Jersey City, New Jersey, United States
August 10, 2018

Contact this candidate




** ***** ***** *****, ****** City, New Jersey, 07310


Stony Brook University 08/2016-05/2018

Applied Mathematics and Statistics, M.Sc.,

Cumulative GPA: 3.3/4

Beijing University of Posts and Telecommunications 08/2011-05/2015 Mathematics and Applied Mathematics, B.Sc.,

Cumulative GPA: 3.5/4 Undergraduate Scholarship Student (Top 20%) COURSES: Regression Theory, Data Analysis, Stochastic Models, Categorical Data Analysis, Design of Experiments, Statistical Computing, Introduction to Databases, Machine Learning SKILLS: R, SAS, SPSS, SQL, ACCESS, C++, TABLEAU, EXCEL (VLOOKUP, PIVOT TABLES), MATLAB, PYTHON EXPERIENCE

China Academy of Telecommunication Research of MIIT 02/2016-07/2016 TeleInfo Institute, Data Analyst Intern Beijing, China

• Performed in-depth statistical analysis in growth-trend of mobile users and 4G users in Shanxi province and proactively identified new insights by exploring the relationship between telecommunications development and economic growth.

• Developed a Least Squares Curve Fitting Algorithm with Numpy in Python to predict the number of mobile users in the future 6 years, which improved the High-Level-Planning index by 73%(Growth Rate Prediction Method) to 89%.

• Estimated a simultaneous equation model and conducted exploratory data analysis and visualization by ggplot2 in R.

• Summarized the report regarding the level of communication of Shanxi province in 2015, which is released on the MIIT official website. Beijing Mercedes-Benz Sales Service Co., Ltd. 08/2015-01/2016 Controlling Department, Intern Beijing, China

• Consolidated and monitored all Service Level Agreements with related parties, measures, analysis effects.

• Identified risks and chances against the target and proposed potential measures to Senior Manager and CFO to secure budgets target.

• Prepared budget forecast based on spending trends and up-coming projects by Time Series Procedures (PROC TIMESERIES, PROC FORECASTING) in SAS, reviewed and proposed update annual budget. Mercedes-Benz Auto Finance Ltd. 02/2015-07/2015

IT Management, Intern Beijing, China

• Supported project management, tracked project status, found potential bugs in data tracking, coordinated with different parties for system function go-live.

• Effectively generated and communicated analysis and insights on application performance to the business team.

• Implemented, maintained and developed data fronted (Apps & Desktop) with vendors to provide users with a simple and friendly portal for their reporting and analysis needs.

• Created SQL statements to extract data needed for the report.

• Assisted in the implement and maintenance of the Data Warehouse as the single source of truth for data and continuously worked on expanding data scope and integration.


Statistical Computing Class Project 02/2018-05/2018

• Built an R package implementing the first order Hidden Markov Model algorithm by using the Forward-Backward algorithm and Viterbi algorithm to decode a sequence of observations which is decided by users.

• Conducted Bayesian Information Criterion to simulate the number of hidden states which is unknown.

• Solved the HMM estimation by using Expectation-Maximization framework. Design and Analysis of Experiment Project 09/2017-12/2017

• Designed a two-level fractional factorial experiment and determined the potential function by requesting the values of the dependent variables generated by the model for a specified set of runs.

• Designed and Analyzed the data by using FrF2 and DoE in R.

• Checked the model adequacy by Shapiro-Wilk test and Box-Cox transformation. The result showed that our estimated model had predicted all the correct variables as the real model given by TA. Netflix User Rating Prediction 09/2017-12/2017

• Applied Singular Value Decomposition Method for collaborative filtering on Netflix Prize dataset and showed the performance with the Root Mean Squared Error is 0.964.

• Worked on the predictors, performance, diagnose and built the framework on C++.

• Improved and limited overfitting through Cross Validation. Optimal Demand Response Scheme in AC Power Network 09/2017-12/2017

• Developed a formulated demand response problem as an optimal power flow problem subject to a series of constraints.

• Developed a fully distributed algorithm for resolving the OPF problem.

• Simulated the proposed DR model by using MATLAB in the modified IEEE standard distribution system.

• The results showed the convergence of the algorithm and the effectiveness of keeping demand under limit. Graduation Project 12/2014-05/2015

• Built a weighted Naïve Bayes Classification model based on partial least squares regression.

• Collected Dissolved gas-in-oil analysis data and electrical tests as the necessary attributes to classify the fault type of transformer.

• Proved that the new model avoided the influence from unrelated variables and got high efficiency to diagnose the transformer fault samples. Prediction error of the final model was about 10% of true values of diagnose efficiency, outperformed the results of 2013 graduates.

Contact this candidate