** Manage monthly data updating process for company database, develop quality assurance (QA) materials and improve predictive algorithm.** Complete predictive analytics projects to help various clients understand market evolution, often resulting in repeat business requiring expanded analyses in other regions or for other products. Create quantitative models to predict market demand of products, R Markdown analysis reports for better internal communication/interpretation, and user-friendly R scripts enabling analysts to run R code. Explore various methods, including linear regression, panel analysis, principal component analysis, and Monte Carlo simulation, select models with good predictive power and statistical fitness with easy-to-communicate interpretation. Present complex statistical methods to non-statistical audiences in easy-to-understand language.** Developed method to create % of households within certain income bracket. Collect and clean raw data.** Worked on internal project to improve company’s renewal rate. Collected wide range of data from Salesforce, analyzed correlation of variables against renewal decision, and built logistic regression model to access risk of current client based on variable values.** Created interactive and user-friendly web application with R Shiny for analysts to access monthly/quarterly data. Allow users to select indicators, markets, and time periods and include download functionality.

** Played key role in building company database containing thousands of macro/microeconomic indicators. Wrote web scrapers in Python for public data sources to extract data and helped design ETL pipeline, developed/implemented time series predictive algorithm achieving high predictive accuracy, and maintained MySQL database storing metadata.** Ran large amount of regression models with R language in Amazon’s Elastic MapReduce (EMR).

** Assessed feasibility of building company database to save money from data purchases and avoid contracting issues. Identified potential public data sources for extraction and analyzed possible challenges.

** Performed statistical analysis in R for high-performance computer system, extracting useful statistics from large usage profile and operation information dataset and improving annual usage report with enhanced graphs and charts.** Presented work to scientists at an exhibition.