A Fine Slice Of SQL Server
Nina Zumel takes us through Cohen’s D, a useful tool for determining effect sizes in experiments:
Cohen’s d is a measure of effect size for the difference of two means that takes the variance of the population into account. It’s defined as
d = | μ1 – μ2 | / σpooled
where σpooled is the pooled standard deviation over both cohorts.
Read the whole thing.
Comments closed Ulrik Stervbo has a performance comparison for for, apply, and map functions in R:
It is usually said, that for– and while-loops should be avoided in R. I was curious about just how the different alternatives compare in terms of speed.
The first loop is perhaps the worst I can think of – the return vector is initialized without type and length so that the memory is constantly being allocated.
The performance of map isn’t great, though the benefits to me are less about performance and more about readability. H/T R-bloggers
Anisa Dhana shows how you can create a quick heatmap plot in R:
To give your own colors use the
scale_fill_gradientnfunction.ggplot(dat, aes(Age, Race)) +geom_raster(aes(fill = BMI)) +scale_fill_gradientn(colours=c("white", "red"))
This is a quick example using ggplot2 but there are other heatmap libraries available too.
Bruno Rodrigues shows one technique for forecasting intermittent data:
Now, it is clear that this will be tricky to forecast. There is no discernible pattern, no trend, no seasonality… nothing that would make it “easy” for a model to learn how to forecast such data.
This is typical intermittent demand data. Specific methods have been developed to forecast such data, the most well-known being Croston, as detailed in this paper. A function to estimate such models is available in the
{tsintermittent}package, written by Nikolaos Kourentzes who also wrote another package,{nnfor}, which uses Neural Networks to forecast time series data. I am going to use both to try to forecast the intermittent demand for the{RDieHarder}package for the year 2019.
Read the whole thing. H/T R-Bloggers
Comments closedAndy Kirk shares the results of a rather informal Twitter poll:
Yesterday I ran a simple Twitter poll about the relative ease of learning R vs. Python. Although a correct answer to this query will ALWAYS have to be based on nuances like pre-existing skills and the scope of need, this originates from people telling me they encounter job or career profiles that list a need for R and/or Python. If they don’t have either, if they prioritised the pursuit of just one, which would be possible to develop a degree of competency more easily, more quickly and more efficiently?
Andy has also created a Twitter moment from the responses.
My thought, based only on the question itself, is that R would be better than Python because the hypothetical person has no additional programming skills. For someone with additional programming skills, the breakdown for me starts with, if your background is statistics, database development, or functional programming, you probably want R; if your background is object-oriented development or imperative programming, you probably want Python. And then it gets nuanced.
Comments closedCapri Granville explains hexagonal binning to us and gives a few examples:
The reason for using hexagons is that it is still pretty simple, and when you rotate the chart by 60 degrees (or a multiple of 60 degrees) you still get the same visualization. For squares, rotations of 60 degrees don’t work, only multiples of 90 degrees work. Is it possible to find a tessellation such that smaller rotations, say 45 or 30 degrees, leave the chart unchanged? The answer is no. Octogonal tessellations don’t really exist, so the hexagon is an optimum.
Every time I see one of these, I think of old-timey strategy war games.
Comments closedKristian Larsen shows off some linear algebra concepts in R:
In this article, you learn how to do linear algebra in R. In particular, I will discuss how to create a matrix in R, Element-wise operations in R, Basic Matrix Operations in R, How to Combine Matrices in R, Creating Means and Sums in R and Advanced Matrix Operations in R.
The post is so chock-full of examples, the only block of multi-line text is the description.
Comments closedWilliam Doane has published a sample RStudio configuration:
Whenever I need to install RStudio on a new machine, I have to think a bit about the configuration options I’ve tweaked. Invariably, I miss a checkbox that leaves me with slightly different RStudio behavior on each system. This post includes screenshots of my RStudio configuration and custom keyboard shortcuts for RStudio 1.3, MacOS, so that I have a reference.
I like these kinds of posts because they can help you find interesting settings you might not otherwise know about. Also, I second the FiraCode recommendation for R as well as F#. The only reason I don’t use it more is because I don’t want to confuse people during presentations. H/T R-Bloggers
Comments closedDave Mason takes us through showing an R-based visual in Power BI:
The R engine isn’t included with the installation of Power BI desktop. I won’t go into detail on this, so just know you’d need to install that separately. I had already installed the R component as part of Machine Learning Services for SQL Server 2017. I also had RStudio installed. Within Power BI desktop, take a moment to click File | Options and settings | Options to open the Options page. Then click R scripting in the list of Global Options. Here you’ll see options to set the R home directory and the desired R IDE.
Click through for the demo.
Comments closedTomaz Kastrun takes us through the structure() function in R:
Structure() function is a simple, yet powerful function that describes a given object with given attributes. It is part of base R language library, so there is no need to load any additional library. And also, since the function was part of S-Language, it is in the base library from the earlier versions, making it backward or forward compatible.
Read on to see how you can create a matrix or data frame using this function and additional details you can save.
Comments closed