@bthall if i'm not mistaken, this already exists. for example, if you're in a julia shell, you can type

?≥

and you get this:

help?> ≥
"≥" can be typed by \ge<tab>
search: ≥

>=(x, y)
≥(x,y)

Greater-than-or-equals comparison operator. Falls back to y <= x.

Examples
≡≡≡≡≡≡≡≡≡≡

julia> 'a' >= 'b'
false

julia> 7 ≥ 7 ≥ 3
true

julia> "abc" ≥ "abc"
true

julia> 5 >= 3
true


: Docstrings for math functions/operations. With you can use Unicode, so you could even use general purpose symbols for function names.

@bthall I would try visualising the data points from both perspectives. I've found RapidMiner a good software (non free for productive purposes) for doing this. It allows you to apply most data science algorithms, visualize them and find connections or new data points. Once I find a good representation of my data, I can move the actual implementation to something more efficient (R, Numpy, scikit, etc)

For people here familiar with or , have you heard of methods that bring together the two approaches?

For analyzing my wiki's contents I could analyze the explicit linkages between articles with graph theory and the implicit linkages with text mining, but I wonder about how much better it'd be to analyze both graphs. I think that in textmining creates a graph representation anyway, so it may just require tools for integrating two related graphs.

@RaoOfPhysics Sure, just switch over to game studies.

Agricultural Econ / Sociology Question

I'm current tinkering with the #Wallabag #android app to see if I can get #MathJax rendering working in it to nicely render mathematical equations written in #LaTeX. So far I have the base script running on my phone, but I didn't realize that I need to load in a bunch of additional files to get it to work, so I'm working on getting the additional files loaded in. Not bad progress for a guy that barely understands #AndroidStudio. :D

Scientific conference poster sessions: what you need to do.
Step 1: Remember to bring your poster with you.
Step 2: Make sure it's securely fixed to the board.
Step 3: Stand nearby in case someone has questions.

I have been experiencing an inverse Dunning-Kruger effect over the last few years. I have realized that literally everyone is better than me at something, and so I am much better off collaborating with them than trying to work on my own.

@bthall In broad strokes, yes, but the devil is in the details.

I suspect that I've run into a situation where designing and running simulations is relevant.

I came up with a basic formula that I think might provide a simple way of inferring something about a statistical distribution, but it'd be nice to produce a way of testing my idea against generated distributions. :)

Do any of you have experience with this manner of thinking? Is this indeed what people do when they use simulations?

what tools do you use for personal academic 'project management'? tracking status of papers/studies, stuff to read/prioritize reading, tracking cfps and conferences of interest, various sizes of to-do lists...? one tool or a toolkit of varied apps?
(doesn't necessarily need to be shareable, but some parts could extend to co-author communication)

Oh! This is a nice paper summarizing a lot of information about prior choice in Bayesian models. It's targeted at ecologists and the kinds of models they tend to use, but useful outside of ecology I expect.

onlinelibrary.wiley.com/doi/ab

@bthall @invaderxan Do you have any good suggestions for how we *should* organise such papers as pdfs?

Word of caution. If you, like me, save PDF copies of papers to a directory called "to read" or "unsorted" or similar, try not to let them accumulate.

This advice brought to you by that one scientist who just spent literal hours sorting through a folder full of several hundred papers.

It'd be kinda cool if a statistical programming language / package provided checks on whether one's use of methods is theoretically proper or not.

Recently I managed to realize that I was planning to plot some of my personal finance data in a way that would suggest that path dependence isn't playing an important role in it, when it quite likely is.

I was checking some of the seeds I'm germinating, and look! This orange seed is polyembryonic!

Polyembryonic seeds are basically what happens when a plant has twins, and one seed contains more than one embryo. Some plants (like some types of mango) always grow seeds like this. In my experience, it's unusual in oranges though.

Last night I gave a short presentation at a meetup, and it went pretty well. I demonstrated that you can easily use interactive widgets in #Jupyter notebooks to explore a problem and present the notebook in a way that is useful and less worrying to code-averse users (hides code, retains text and interactivity via the "appmode" extension). I showed this with #Python but included that this works in #Julia, too.

You can see the presentation notebook here: https://gitlab.com/bthall/econ-notebooks/blob/master/PayoutAnnuity%20-%20Talk%20Copy.ipynb

@bthall If you want to stick with python and still do Rethinking Statistics, you can! github.com/pymc-devs/resources