R, Python, JMP, SQL, Tableau, Power BI, DecisionPro, SAS, Google Analy

Xinyu “April” Liang

+1-213-***-**** ● ac3h4y@r.postjobfree.com

EDUCATION

University of Southern California, Marshall School of Business – Los Angeles, CA Dec 2017 Master of Business Analytics

Nankai University (NKU), School of Economics – Tianjin, China June 2016 B.S., Public Finance Minor: Physics (top 10% of cohort) Honors: Reported by major media (e.g. Xinhua Daily) because of academic excellence and extra-curriculars SKILLS

Data Analysis & Business Intelligence Tools: R (ggplot2, ggmap, plotly, rvest, dplyr, lubridate, rshiny), Python (numpy, pandas, pyspark), JMP, SQL, Tableau, Power BI, DecisionPro (Marketing-specialized), SAS, Google Analytics, Excel Web Development: Bluehost, Weebly

Database Construction & Data Warehousing: Oracle, AWS, DBeaver, Postgres, pgAdmin Statistical Modeling: Supervised learning methods (Linear Regression, Logistic regression, KNN, Decision Trees, Ridge, LASSO, Random Forest, Time Series analysis) Unsupervised learning methods (Kmeans, Hierarchical Clustering) BUSINESS EXPERIENCE

City of Los Angeles, Controller’s Office – Los Angeles, CA Summer, Fall 2017 Data Analyst Intern

• Developed 9 cutting-edge automated tools of Payroll, Personnel, Budget Monitor & Projector that projected future expenditure ( 5% error); resulted in adoption by executives with a $2 Million & 95% data process time saving for the City per dashboard; built, designed and published 4 APPs to the entire organization for internal use;

• Collaborated with ITA Department for BI Infrastructure evolution (automated data refreshing scheduler, i.e. gateway configuration, database connection, and cloud service setup); performed and optimized data ETL in SQL from Oracle Database; Trained new interns for effective PowerBI, and Oracle Database use SixThirty Group – Los Angeles, CA 2016 - 2017

Data Analyst

• Predicted crime rate in Santa Monica using time series approach to evaluate the impact of the newly-opened Metro line, identified abnormal area and recommended to strengthen patrols in blocks near metro station;

• Performed exploratory analysis of water usage data (01/2008 – 01/2017), evaluated the significance of Water Usage Allowance Policy, provided recommendations for water saving, e.g. grant recognition for “Government - Certified Water - Saving house” to boost resale value PricewaterhouseCoopers, LLP – Beijing, China Summer, 2015 Marketing Analyst Intern, Dept. of Assurance Audit

• Assisted China Financial Services Leader and Partners on market intelligence research & analysis for business development and other purposes; examined return on investment for diversified advertising campaigns Agricultural Bank of China – Dongying, China Winter, 2014 Credit Analyst Intern, Dept. of Credit Management

• Analyzed investment strategies for incoming clients, contributed to the decision-making process of credit authorization through successful identification of potential fraud behavior ANALYTICS EXPERIENCE

Choice-Based Marketing Conjoint Analysis for Media Streaming Device, Xiaomi Inc.

• Created a conjoint survey with demographic and choice-based questions to observe target respondents

• Introduced a new feature “Asian Channel”, and analyzed the part-worth for all attributes (e.g. price, brand, etc.), estimated the willingness to pay for Asian Channel

• Implemented Chain Ratio Method and Pricing Strategy for market size assessment, recommended the optimal price for new-added feature to maintain market share with profit margin retained or improved Online Article Popularity Contributor Analysis (Machine Learning & Statistical Methods)

• Identified and quantified the key features (word polarity, release day, etc.) that prompt an article popular by implementing shrinkage methods (Lasso and Ridge), machine learning algorithms (Linear & Logistic Regression, bagging, LDA, KNN, Random Forest and Decision Tree, with achievement of 75% prediction accuracy University Student Innovation Non-Profit Project Campaign; 1st Price and Team Lead

• Constructed logistic regression for predictive analysis of feedback based on multi-dimensional segmentation (age, income, districts); combined discriminant analysis with classification via Hierarchical and K-Means approach to determine age group (26.2 – 34.7) clusters for the policy promotion

Contact this candidate