Tested the key elements for basketball players’ overall value, a general ability, in the NBA 2K16.
Collected big data and cleaned it with Excel and SQL, fitted multiple linear model in R for NBA player’s overall value, checked the effect of coefficients by doing Hypothesis Testing, and got new qualities for players by Principal Component Analysis.
CONCLUSION: Select five principal components to reflect five different capabilities, including defense, attacking and score, comprehensive, breakthrough, and basic.
Researched the most important elements for Chinese overseas students and guided them to get jobs in USA.
Designed survey to collect data, cleaned data by Excel, SQL or Oracle, and fitted the initial model by Linear Regression, then did cross-validation to check the accuracy of the model, explained the data via the result getting from Principal Component Analysis and Classification.
CONCLUSION: 3 variables are important to job offers: GPA, time living in English Speaking Countries, and the number of job interviews.
Analyzed big data, tested the effect of government policy on controlling the high concentration of PM 2.5.
Cleaned data by Excel (get daily average PM2.5 value by self-coding Excel function) and SQL, built ARIMA model, forecasted the concentration of PM2.5 from Aug to Dec.2016, and used CCF to evaluate the effect from North China to South China.
CONCLUSION: click a self-built WEEBLY website (http://group4pm25.weebly.com/). GuangZhou performs best among 5 cities; the effect of PM2.5 from BeiJing to ShangHai is reduced after 2013; the policies work well on controlling PM2.5.
Determined that if there is age discrimination for Boeing in the rehiring process when a lower percentage of older workers than younger ones were rehired.
Used 40 years old to divide the employees to young and old. Used Fisher’s Test to exam the significance of the association (contingency) between the two kinds of classification; selected Breslow-Day Test to check the assumption, homogeneous association of Apsley vs Boeing dataset, for Cochran–Mantel–Haenszel test; selected Cochran–Mantel–Haenszel test to check the association in stratified data, 19 units in Apsley vs Boeing case.
CONCLUSION: Older workers were easy to be fired than younger ones. There is age discrimination for Boeing in the rehiring process.
Found key factors that influence the survival rate, evaluated the efficacy of hormone therapy and offered advice based on research.
Used Excel function and SQL to clean data and stepwise cox model selection to fit the best model, did regression diagnostic to determine the kept variables and analyzed 3 problems via stratified proportional hazard model.
CONCLUSION: tumor size and so on will reduce survival rate; patients, who are pre-menopause over 55 years old, have few chances to get cured; hormone therapy is recommended for receptor-test resulting to positive patients.