Press "Enter" to skip to content

Day: November 15, 2021

Replacing p-values with Bootstrapped Confidence Intervals

Published 2021-11-15 by Kevin Feasel

Florent Buisson has an interesting post on avoiding p-value calculations:

And indeed, I worked with highly-skilled data scientists who had a very sharp understanding of statistics. But after years of designing and analyzing experiments, I grew dissatisfied with the way we communicated results to decision-makers. I felt that the over-reliance on p-values led to sub-optimal decisions. After talking to colleagues in other companies, I realized that this was a broader problem, and I set up to write a guide to better data analysis. In this article, I’ll present one of the biggest recommendations of the book, which is to ditch p-values and use Bootstrap confidence intervals instead.

I’m a committed Bayesian (or at least a Bayesian who should be committed—depends on who you ask), so I’d consider this a big step forward.

Comments closed

Monotonic Constraints on Random Forests

Published 2021-11-15 by Kevin Feasel

Michael Mayer has some interesting R and Python code for us:

On ML competition platforms like Kaggle, complex and unintuitively behaving models dominate. In this respect, reality is completely different. There, the majority of models do not serve as pure prediction machines but rather as fruitful source of information. Furthermore, even if used as prediction machine, the users of the models might expect a certain degree of consistency when “playing” with input values.

A classic example are statistical house appraisal models. An additional bathroom or an additional square foot of ground area is expected to raise the appraisal, everything else being fixed (ceteris paribus). The user might lose trust in the model if the opposite happens.

One way to enforce such consistency is to monitor the signs of coefficients of a linear regression model. Another useful strategy is to impose monotonicity constraints on selected model effects.

Certain types of regression algorithm make this easy, but random forest? Not so much. That’s where Michael steps in.

Comments closed

Creating a Clock with Powershell

Published 2021-11-15 by Kevin Feasel

Jeffery Hicks knows what time it is (it’s clobberin’ time):

I’ve published a new project to the PowerShell Gallery. This is something that I needed, and maybe you do as well. Even though I have the typical clock running in the Windows taskbar, I have an ultrawide monitor so it isn’t always easy to read. I had been running the xclock app from WSL which was nice. But realized what I really wanted was a WPF-based clock that would display a formatted date-time string on my desktop.

Now there’s a PSClock similar to xclock, and you can check it out.

Comments closed

Using ConcatenateX in Power BI

Published 2021-11-15 by Kevin Feasel

Reza Rad describes a DAX function:

It happens often in Power BI calculations and reports that you need to concatenate a list of values from a column. You can do this concatenation in Power Query or DAX. However, if the concatenation needs to be done dynamically. ConcatenateX is a very helpful DAX function to achieve such results. It is very helpful to understand what happens in the virtual tables in DAX too. In this article and video, I’ll explain what ConcatenateX is and how it works in Power BI and DAX.

Click through for a video, as well as a detailed explanation in blog post format.

Comments closed

Pokey Performance with EXISTS

Published 2021-11-15 by Kevin Feasel

Erik Darling reminds us that even good things can go bad:

Look, I really like EXISTS and NOT EXISTS. I do. They solve a lot of problems.

This post isn’t a criticism of them at all, nor do I want you to stop using them. I would encourage you to use them more, probably.

But there’s some stuff you need to be aware of when you use them, whether it’s in control-flow logic, or in queries.

But do read on to see a specific type of issue you can run into with a left semi join.

Comments closed

Dynamic Column Rename in Power BI with XMLA and TOM

Published 2021-11-15 by Kevin Feasel

Kristyna Hughes solves a problem:

For the TOM and XMLA experts, imagine this. Your customer wants to dynamically rename columns without using the Power BI Desktop and would prefer all existing report visuals not get broken by the new name. Impossible? Not with TOM, XMLA, and translations within Power BI.

If you’ve ever tried to change a column name in a Power BI source, you’ve likely run into this error on any visuals that contained the renamed column. And when you hit that “See Details”, it will tell you the column that you simply renamed is no longer available for your visual.

So how do we get around that?

Read on to see how.

Comments closed