Linda (Mengyu) Zhang

• Retrieved companies’ credit rating data from Standard & Poor’s and did multi-classification on it • Predicted new companies’ credit ratings using some classification models such as principal component analysis, logistic regression, naïve bayes method, support vector machine and random forest method

• To form a general overview of the realty industry in Hangzhou, China by collecting related data and analyzing relevant mathematics models. This could also provide reference to realty investment. • focused on theories of time series, financial mathematics and financial risk management course and layed the theoretical foundation in forms of group study; mastered data analysis software such as C language, R language and so on;

This wine data is from UCI Machine Learning Repository and it contains two datasets. The two datasets are related to red and white variants of the Portuguese "Vinho Verde" wine. There are 12 variables and 4898 observations of white wine, 1599 of red wine. Among all variables, fixed acidity, volatile acidity and citric acid are predictors measuring the degree of acidity. Residual sugar, chlorides measure other indexes. Free sulfur dioxide and total sulfur dioxide are measuring the sulfur dioxide’s content. These are all chemical predictors. Density, pH, sulphates and alcohol are measuring physical attributes. The quality of wine is with the score between 0 and 10. For my individual part of this project, I used the discriminant analysis to classify the quality of red and white wine. I classified the data into six levels that are red wine with low quality (RL), red wine with median quality (RM), red wine with high quality (RH), white wine with low quality (WL), white wine with median quality (WM) and white wine with high quality (WH). Classifying a new data point into one of the groups above based on the discrimination is my final goal of this project.

Provided with historical sales data for 45 stores in 81 departments of Walmart in different region, I predicted the weekly sales for recent years using different methods of regression model such as simple linear model, logistic regression model and random forest model. With this prediction, I can see how time, holiday and other possible factors will affect the future sales.

Activities and societies: Jasmine field band of CSSA Statistics Club

Activities and societies: Violinist at Wenqin Symphony Orchestra , Vice director of Student Union Department of Mathematics