In the last post I presented a function for recovering marginal effects of interaction terms. Here we implement the function with simulated data and plot the results using ggplot2. #---Simulate Data and Fit a linear model with an...

Couple of months back I read Jeffrey Breen’s presentation on mining Twitter for consumer attitudes towards airlines, so I was just curious how it would look if I estimate the sentiment toward major hotels. So here it is: # load twitter library > library(twitteR) # search for all the hilton tweets > hilton.tweets=searchTwitter('@hilton',n=1500) > length(hilton.tweets)

Firstly, I know nothing about election fraud – this isn’t a serious post. But, I do like to do some simple coding. Ben Goldacre posted on using Benford’s Law to look for evidence of Russian election fraud. Then Richie Cotton did the same, but using R. Commenters on both sites suggested that as the data

For the longest time I resisted customizing R for my particular environment. My philosophy has been that each R script for each separate analysis I do should be self contained such that I can rerun the script from top to bottom on any machine and get the same results. This being said, I have now

While playing around with Bayesian methods for random effects models, it occured to me that inverse-Wishart priors can really bite you in the bum. Inverse Wishart-priors are popular priors over covariance functions. People like them priors because they are conjugate to a Gaussian likelihood, i.e, if you have data with each : so that the