We have an XBOX360 at work, and the game that is played the most often is definitely NHL2013. Since I’m a professional data nerd, I wanted to see how I compared to the other players. As such, I wrote a simple Rails web application to track games and thus rankings/stats of everyone who plays at work.
It definitely needs a lot more work, but I’ll be adding to it over time. It just works for now.
Check it out on GitHub here.
Imagine this: you are travelling from one city to another. You absolutely hate taking the bus, but you find it too expensive to rent a car. Wouldn’t it be nice if you could somehow mix the convenience of driving with the economy of taking a bus? Maybe you can.
One thing that becomes quite apparent when estimating the prices of cars is that different cities have fairly different prices, sometimes starkly different. For example, cars in Calgary are much more expensive than cars in Ottawa or Toronto. However, due to some basic economics theories on arbitrage and the cost of travelling between the cities, we would think this would be close to impossible to turn a profit from arbitrage. I aim to prove that it is feasible to buy a car in Ottawa or Toronto and sell it in Calgary for profit.
A friend referred me to an article on using big data to find various correlations in the real world. One such interesting correlation is that orange cars tend to be less prone to breakdowns. I thought it would be interesting to attempt to recreate this with the data set that powers Carbitrage, so here I go.
I’m currently working on a bookmarklet for Carbitrage and Kijiji integration. If you are searching for a new car on Kijiji and you stumble upon one of interest, you can click the bookmarklet, which will direct you to a page with the estimated market price for that specific car. It also calculates how over- or under-priced that car is.
I’m happy to announce my new website, Carbitrage. This website, once complete, will offer users the ability to get an estimate on the market price of a used car or truck. It is currently in testing, but I’m estimating that it will take about two weeks from now until it is ready for public use.