Statistics treatment for a survey study designed to test the hypothesis: If the means of solar energy generation are more aesthetically pleasing consumers will be more willing to convert to solar energy.

A worksheet for exploring common R Language capabilities via common R vocabulary. While taking STAT290 at Stanford University, this was inspired by Chapter 4 of Hadley Wickham’s book Advanced R http://adv-r.had.co.nz/Vocabulary.html.

We set up a test to detect comparative measurement bias between two Fitbit devices worn on the same canine subject. Descriptive statistics and some inferential testing address questions about variability of measurement and whether a significant measurement bias is found.

An update of the worksheet investigating activity metrics calculated from raw data collected by Fitbit wearable devices on canine whelps (puppies). This version digs into the question of measurement variability between the two Fitbit devices, to see if a significant measurement bias exists.

How my dive into this Kaggle contest taught me to think about bias-variance tradeoff in a sense beyond the data in front of me.

A working implementation of the Kaggle sample solution knn_benchmark.R which I've submitted to Kaggle.com for team Winter14 in the Digit Recognizer contest. This was our first submission, and curiously it ranked 214 out of 360 at the time of submission even though it is a simple baseline solution based on sample code from Kaggle.

A worksheet to investigate activity metrics calculated from raw data collected by Fitbit wearable devices. The twist is that this study is not on humans - it's on puppies. I'm working with my daughter on her study "Whelp Sleeping and Waking Activity Patterns as a Predictor of Service Dog Success."