I keep talking about using large datasets and how some models take quite long time to run. One of those models is typically two way fixed effects models. While regular econometrics packages like plm are quite good for reasonably large datasets, more and more I find it hard to use with my datasets. Here is where LFE comes in. Developed by Dr. Simen Gaure, this package demeans the data in two dimensions (typically the demeaning only happens in one dimension).
This semester I will be serving in OU COB Research Colloquium. We organize research presentations in college of business. As part of my duties I was tasked with creating posters. I considered my options and decided that creating new posters every other week was too much work.
So I created a tex file that grabs the relevant information with datatool package and uses a beamer template to generate the poster.
I am doing a discrete choice experiment design for a project that we are just starting. The idea is to present screenshots of various app’s in an app store and ask the participants to pick one. For this, we need screenshots of app’s with key variables changed. The problem is that, unless you are careful you may make mistakes in creating manipulations. Especially if there are more than a handful conditions. For this, I used GIMP and xcftools package to programmatically create the treatments.
Tonight I finally finished Last Night on Earth miniatures. I painted the heroes a while back. For a while I was pretty satisfied with the effort. Then I saw Sorastro’s Zombicide painting guide for zombies… Having painted the zombies I thought I was done, but then I saw Consadine’s basic basing guide. Well finally tonight I think I am done for good with LNoE (maybe except for a foam core box insert).
Azade is old enough to play simple games (Animal upon Animal, Brandon the Brave) but dice throwing has been a problem. A while back I decided to build a dice tower with the dual purpose of both playing games with kids and to make sure the dice is properly random (no cheating). Continue reading “Foamcore Dice Tower”
I was looking to visualize some interactions for our engagement study. I decided to use a 3D plot but the R base plots did not look appealing. I guess ggplot2 pampered me a bit. My search lead me to plotly. Visually pleasing plots with R and Python.
I was in Hawaii last week to present “If You Let Them Build It, They Will Stay! An Empirical Study of Add-on Content and User Engagement”. It was a great honor to see the paper nominated for best paper award, alas we did not receive the award. This is the first paper we released based on the Steam Dataset I have been collecting and I am very excited to see where this dataset will take us.
In Hawaii for HICSS 50. So far so good. Have seen old friends, met new friends and am enjoying Panoply security competition. Paper being nominated as best paper does not hurt either.
Alright, so far I have been posting about my research and teaching… Now that I have a job that I feel fairly confident about, I think I can actually blog with a bit more liberty. I have been a huge fan of boardgames since I laid my eyes on the “Milyoner” (Turkish knockoff of monopoly). Moving to the US has only broadened my horizons when it comes to board games and now I have a decent collection of my own.
Recently I ascended to a new level in the hobby. I am now painting the miniatures that come with some of my games. I must admit I am pretty new to the hobby and am not implementing all the tips and tricks a seasoned painter knows. Continue reading “Board Games and Miniature Painting”