Author

Christine Garcia

Browsing

When we think of computers, we think of the twenty-first century. But did you know that India started using them back in the 1950s? Computers were the unexpected secret sauce for accelerating India’s economy after Independence. The government used them to process massive amounts of data and generate complex statistical models of the economy. How did the Indian government start…

Landsat is without a doubt one of the best sources of free satellite data today. Managed by NASA and the United States Geological Survey, the Landsat satellites have been capturing multi-spectral imagery for over 40 years. The latest satellite, Landsat 8, orbits the Earth every 16 days and captures more than 700 satellite images per day across 9 spectral bands…

Today hedge funds use alternative data to shape their investments. Why can’t organizations use it to shape their own future? In the fall of 2015, Chipotle (an American food chain famous for its massive burritos) was rocked by a series of health scares. From August to December, norovirus and salmonella broke out in 19 locations across California, Massachusetts and Minnesota. The…

Today’s news is filled with stories of data misuse — for example, Cambridge Analytica harvesting personal data via Facebook and Amazon, Facebook’s Android app logging metadata on every call and text, and Grindr revealing its users’ HIV status to third parties. These companies are being roundly criticized for misusing their users’ data. In the midst of this uproar, companies are…

What is an outlier? In short, it’s a data point that is significantly different from other data points in a data set. The long story? There isn’t a strong mathematical definition for what is or isn’t an outlier. In the end, detecting and handling outliers is often a somewhat subjective exercise. So how can you dive into a new data…

Nothing is more frustrating than wrapping up a lengthy data collection exercise, aggregating all the data and looking through it, only to find missing data. At best, these missing values are a nuisance that can be fixed with a bit of work. At worst, they pose an intimidating threat to data quality and your sample size. How can you assess…