demo

Flexdashboard, Dygraph lung deaths

Make xy charts and a box plot

@ThomasBrand created the original version of this in Macroeconomics Observatory - R. It appeared in R-bloggers on June 2, 2020

This code uses data provided by MakeoverMonday January 2018 to identify flight speed of turkey vultures during migration.

This is a reverse engineering of a Unsupervised Learning and Text Mining of Emotion Words in R published in Data Science on May 17, 2017. While the original used Warren Buffet letters to shareholders, this exercise uses lyrics for the 18 songs written by Jimmy Buffet. Prepared in July 2017 to share with the Buffalo R Users Group at Roswell Park

5 page summary of the project, model, app

Text Prediction Model

Week 2 assignment to summarize text data for Swiftkey project

An interactive app that lets you pick flowers

An interactive app that lets you pick flowers

The final project for Coursera Data Products, Data Science Specialization with Johns Hopkins University

Here is a shiny app that makes it easy for an analyst to toggle between frequency data and a chart of the overall distribution of survival data.

A plotly chart using beaver1 and beaver2 data

A leaflet exercise

A linear modeling analysis of the mtcars data to determine the impact that transmission has on mpg

This document uses simulations of exponential variables to demonstrate the Law of Large Numbers and the Central Limit Theorem. This is for assignment #4 in Coursera, Statistical Inference

This analysis explores the NOAA Storm Database from 1950 to 2011 to determine 1) acrosss the United States, which weather events are the most harmful to the health of the population and 2) which have the greatest economic impact. Tornados and heat emerge with the highest cumulative fatalities, while hurricanes and ice storms generate the greatest economic damage. The analysis first identifies all weather events that resulted in at least one fatality, cleans that subset so it aligns with the standard NOAA weather event coding, and then summarizes the events in the 90th percentile. The analysis of the economic impact follows a similar process, with an additional step that combines the property damage cost with crop damage cost in $ billions. The summary of the economic impact focuses on the events in the 95th percentile of total damage cost in billions of dollars. The biggest challenge for both analyses is the data cleaning. The raw information comes from a variety of sources including: county, state and federal emergency management officials, local law enforcement, skywarn spotters, NWS damage surveys, newspaper clipping services, the insurance industry and the general public. As a result the coding of event types varies wildly from NOAA standards and needs reconciling.

Reproducible Research - Coursera Data Science Week 2 assignmen