*** ***** ******, *** ***, Vernon, Connecticut, 06066 linkedin.com/in/shuai-hao-036412129
University of Connecticut Storrs, CT
Master of Science: Statistics Sep.2016-May.2108
Sichuan University Chengdu, China
Bachelor of Science: Finance Engineering Sep.2012-Jul.2016 WORK EXPERIENCE
Harbin Branch, Bank of China Harbin, China
Intern at the Business Department Jul.2015-Aug.2015
• Implemented the basic credit operations and international settlement business in the bank
• Participated in the loan business projects, visited enterprises with the supervisor which enhanced the relationship with enterprise customers and assist with the preparation of project budget proposals. Analysis the risk of the loan business projects and analysis the financial position based on financial statements. Wrote the financial statement analysis
Prudential (Hong Kong) Jan.2013-Feb.2013
• Provided personalized insurance advice to client
• Assisted in ad-hoc projects with reporting and analysis requests
• Participated in the product exploitation of insurance services and securities lending as well as the follow-up of the product operation performance
Key words: Insurance advice, ad-hoc project
Bootstrap method for Type I right censored data Sep.2017-Dec.2017
• Using bootstrap resampling data to estimate the standard error
• The data is from a study of two headache pain relievers. And the patients are divided into two groups, with each group receiving a different type of pain reliever. The time taken for each patient to report headache relief is recorded.
• Find that the bootstrap resample method is an efficient way to estimate the parameter’s standard error and the hazard ratio of the proportional hazard model.
Key words: Bootstrap resample, SAS, right censored data, survival analysis Predictive Modeling -- Partial Withdrawals for Variable Annuities Jan.2017-May.2017
• The variable annuity partial withdrawal model uses a Generalized Linear Model (GLM) framework to predict total partial withdrawals in given years for a block of policyholders.
• The model is to predict partial withdrawals for each policy exposure year. The model is a transformed policy based frequency and severity model.
• The major innovative aspect of this modeling is that the model captures the individual policy’s historical behavior into the modeling, instead of treating each exposure as an independent contribution to the model. Key words: GLM, partial withdrawals, R
Testing explanation effect of sector based Famma-French factors on stock returns Jan.2017-May.2017
• Regressed each sector portfolio excess return on original FF factors. Regressed each sector portfolio excess return on each sector based FF factors. Those factors are constructed based only on the stocks in that particular sector
• Confirmed that FF factors can explain each sector stock returns well and further improved the model fit by using sector based FF factors, especially in sector healthcare, industry, material, information, consumer utility Key words: Stock return, Famma-French factor, S&P500, R, excel(VBA), Bloomberg Data mining Project-Job Recommendation System Sep.2016-Dec.2016
• Build the job recommend system to recommend job seekers jobs based on their click history when job ads are sent to users with Python and PySpark
• Using TF-IDF (term frequency-inverse document frequency) and the Naïve Bayes Classifier to establish the system. Key words: Machine Learning, Naïve Bayes Classifier, Job recommendation, Python, PySpark SKILLS
• Strong background in SAS, R, Python, PySpark, SQL, Tableau, Matlab, SPSS, STATA
• SAS Certified Base Programmer for SAS 9
• Proficient in machine learning Python libraries such as gradient boosting, NumPy, Pandas, Matplotlib, SciKit-Learn.