Browse Tag by Data

Study & Analysis of Household WiFi Speed

I conducted a balanced three-factor factorial study to test the main effects of obstacles, a number of WiFi users, the distance between WiFi router and WiFi users, as well as the potential effects of interactions among these three variables on WiFi speed. Then I used SAS to analyze the data.

Full Article: PDF

Presentation: PPT

Code: GitHub


Financial Health of United States Banks in 2016

This study reports a k-means clustering analysis investigating patterns of the financial health of all 5,870 banks in the United States in 2016 by using five key variables: net interest margin, return on assets, net charge-offs to loans, tier 1 capital ratio, and total risk-based capital ratio. The results suggest that there are nine groups of banks in terms of their financial health. 1,224 banks are found be in the cluster that represented the most unhealthy institutions, whereas only 19 banks are in the cluster described as most healthy. The banks that failed in the first quarter of 2017 are all found in the first cluster.

Full Article: PDF

Code: GitHub

Slides: PPT

The recipes of data

People visualize data, hear data, but have you ever tried to taste the data? Do you know what is the recipe for the data that you are interested in? Too spicy, salty, or sweet? You may find some ideas about how to build a statistical model when making meals.

My lazy dinner is usually made based on my mood although I am starting to use nutrition data to make dinner. Here are some examples:

P.S. Please feel free to email me if you need the recipes:>