Category: Data Science

Muhammad Touhidul Islam explains what cluster sampling is and why it can be useful:

Cluster sampling is defined as a sampling method where multiple clusters of people are created from a population where they are indicative of homogenous characteristics and have an equal chance of being a part of the sample. In this sampling method, a simple random sample is created from the different clusters in the population. This is a probability sampling procedure.

Click through for a few examples of where this can be useful.

Gavita Regunath and Dan Lantos give an overview of time series algorithms:

Time series forecasting is a data science task that is critical to a variety of activities within any business organisation. Time series forecasting is a useful tool that can help to understand how historical data influences the future. This is done by looking at past data, defining the patterns, and producing short or long-term predictions.

Click through for an overview, as well as ten examples of algorithms you can use for handling time series data.

Ridhima Kumar (re-)introduces us to decile analysis:

Decile analysis was once a popularly used technique, however the convention of teaching and bucketing machine learning problems into either ‘classification’ or ‘Regression’ types, lead people to forget Decile analysis type analyses. I am pretty sure, most freshly minted data scientists would not have even heard of Decile analysis. So, coming back to what is Decile Analysis.

Decile Analysis is used to categorize dataset from highest to lowest values or vice versa. (Based on predicted probabilities)

As obvious from the name, the analysis involves dividing the dataset into ten equal groups. Each group should have the same no. of observations/customers.

It ranks customers in the order from most likely to respond to least likely to respond.

Read on to learn the steps and how this ties with the fact that logistic regression is regression.

Bryan Shalloway has a process for us:

While I no longer use it regularly for the purposes of analysis, I will always have a soft spot in my heart for excel. Furthermore, using a “correct” set of data science tools often requires a bridge. Integrating a rigorous component into a messy spreadsheet based pipeline can be an initial step towards the pipeline or team or organization starting on a path of continuous improvement in their processes. Also, spreadsheets are foundational to many (probably most) BizOps teams and therefore are sometimes unavoidable…

In this post I will walk through a short example and some considerations for when you might decide (perhaps against your preferences) to integrate your work with extant spreadsheets or shadow “pipelines” within your organization.

Click through for Bryan’s thoughts on the topic.

John Cook takes us through the notion of confidence intervals and credible intervals:

Suppose you do N trials of something that can succeed or fail. After your experiment you want to present a point estimate and a confidence interval. Or if you’re a Bayesian, you want to present a posterior mean and a credible interval. The numerical results hardly differ, though the two interpretations differ.

If you got half successes, you will report a confidence interval centered around 0.5. The more unbalanced your results were, the smaller your confidence interval will be. That is, the confidence interval will be smallest if you had no successes and widest if you had half successes.

What can we say about how the width of your confidence varies as a function of your point estimate p? 

Read on to learn that answer.