What to do when your model results are ridiculous?
Experienced analysts and modelers know that “Oh shit!” moment all too well. You inspect the first results from your new model and they are obviously wrong. They are an order of magnitude off from what you expected. Or...
More »An Optimal Valentine’s Date
“You know, Valentine’s Day reminds me of the optimal stopping problem.” Not sure if this phrase was uttered elsewhere, but, this is the sort of water cooler conversations at Lumina HQ. I had a recent conversation...
More »How to update y = m x + b with new data
Abstract: Bayesian linear regression made super easy and intuitive! You arrive in a new city for the first time, and you'll need to consider taking a taxi as one of your options for getting around. Given your prior experience...
More »Analytica 5.2 beta
Analytica 5.2 is now in beta! We invite all users with an active subscription to 5.1 to participate in the beta. Learn more about features we're introducing in 5.2 here, and get the latest beta build here. As someone who really...
More »Testing hypotheses about causation
In 2002, I developed a statistical framework for testing whether your data provides statistically significant support for the hypothesis that A causes B. I published only one conference paper with some colleagues on the idea before moving on to other things,...
More »Efficient estimation of the expected value of information (EVI) using Monte Carlo
The expected value of information (EVI) lets you estimate the value of getting new information that reduces uncertainty. At first blush, it seems paradoxical that you can estimate the value of information about an uncertain quantity X before you...
More »Lumina at INFORMS 2018 in Phoenix
We'll have a have a few events at this year's annual INFORMS meeting, held in Phoenix, AZ this November. In addition to our Lumina Decision Systems booth #56 in the Exhibition Hall (come by and say hello!), we'll...
More »How the strange Cauchy distribution proved useful
On Tuesday I had an interesting exchange with Jorge Muro Arbulú, a professor in Peru, about the Cauchy distribution, which also called the Lorenzian distribution. Unlike most probability distributions you encounter, the mean and variance for strange distribution...
More »Using Python to parse a Shapefile
A shapefile (*.shp) is a binary file format used by Geographic Information Systems (GIS). I used an existing Python library (Fiona) to read and parse a shape file, and then imported that data into Analytica. Here I'll show how...
More »How to call a Python function from Analytica
I found it pretty easy to call a Python function from Analytica using COM automation. The COM integration functionality comes included with the Analytica Enterprise edition. In this blog posting, I'll show you the basics...
More »