Category: Data Science

I decided to test a common narrative:

A common theme among Buffalo Bills fans is the idea that the Bills run too many plays out of shotgun near the opposing team’s goal line, and this is hampering their ability to score points. Instead, these fans argue, they should run from under center, either a direct handoff or a quarterback sneak. If you were to press fans on this, I believe you’d also hear that the Bills are unique, or at least uniquely bad, at running such plays.

I’m going to use the nflfastR package to analyze play-by-play data and see just how well this bit of fan wisdom holds up.

Spoiler alert: it doesn’t.

Comments closed

Steven Sanderson performs a logistic regression:

Logistic regression is a statistical method used for predicting the probability of a binary outcome. It’s a fundamental tool in machine learning and statistics, often employed in various fields such as healthcare, finance, and marketing. We use logistic regression when we want to understand the relationship between one or more independent variables and a binary outcome, which can be “yes/no,” “1/0,” or any two-class distinction.

Click through to learn how to do this.

Comments closed

Steven Sanderson puts a halt to things:

When working with time series data, one common challenge is dealing with non-stationary data. Non-stationary time series can be a headache for analysts, but fear not, because we have a handy tool to make your life easier. Say hello to the auto_stationarize() function from the {healthyR.ts} package.

Read on to learn why you want stationary data for time series analysis and how the auto_stationarize() function works.

Comments closed

Steven Sanderson isn’t just spinning in place:

Before we delve into the ts_adf_test() function, let’s understand the concept behind it. The Augmented Dickey-Fuller (ADF) test is a crucial tool in time series analysis. It’s like the Sherlock Holmes of time series data, helping us detect whether a series is stationary or not. Stationarity is a fundamental assumption in time series modeling because many models work best when applied to stationary data.

So, why “Augmented”? Well, it’s an extension of the original Dickey-Fuller test that accounts for more complex relationships within the time series data.

Click through to see how you can use the ts_adf_test() function to get a better feel for whether a time series is stationary.

Comments closed

Michael Mayer has a new package:

The current version offers:

• H statistics per feature, feature pair, and feature triple
• multivariate predictions at no additional cost
• a convenient API
• other important tools from explainable ML:
  • performance calculations
  • permutation importance (e.g., to select features for calculating H-statistics)
  • partial dependence plots (including grouping, multivariate, multivariable)
  • individual conditional expectations (ICE)
• Case-weights are available for all methods, which is important, e.g., in insurance applications.

Click through for an example of how it works, followed by some simple benchmarking to give you an idea of how it performs compared to similar tools.

Comments closed

Steven Sanderson builds a tree:

Decision trees are a powerful machine learning algorithm that can be used for both classification and regression tasks. They are easy to understand and interpret, and they can be used to build complex models without the need for feature engineering.

Once you have trained a decision tree model, you can use it to make predictions on new data. However, it can also be helpful to plot the decision tree to better understand how it works and to identify any potential problems.

In this blog post, we will show you how to plot decision trees in R using the rpart and rpart.plot packages. We will also provide an extensive example using the iris data set and explain the code blocks in simple to use terms.

Read on to see an example of how to do this.

Comments closed