Random thread

Discussion for all topics (until the forum becomes large enough to justify splitting things up this will be where all topics go)
theojones
Site Admin
Posts: 14
Joined: Mon May 01, 2017 2:18 am

Re: Random thread

Post by theojones » Sat May 13, 2017 5:55 am

Guest wrote:
Sat May 13, 2017 5:50 am
I've been having the same issues for a few days now; for me it definitely didn't start or get noticeably worse yesterday/last night.
Ok. Do you think you could post a screenshot of the issue when it happens next? I'm having a hard time reproducing the problem, so, more information would be quite useful.

chel_of_the_sea
Posts: 27
Joined: Mon May 08, 2017 9:39 am

Re: Random thread

Post by chel_of_the_sea » Sat May 13, 2017 6:45 am

Will do. Firefox was telling me it was an unknown SSL cert issuer, if that helps.
:arrow: TL;DR: 27, trans woman, M.A. in math, Seattle area. Tutor by current trade, but in a bit of professional limbo (if you know anyone hiring, let me know!).

theojones
Site Admin
Posts: 14
Joined: Mon May 01, 2017 2:18 am

Re: Random thread

Post by theojones » Sat May 13, 2017 7:05 am

chel_of_the_sea wrote:
Sat May 13, 2017 6:45 am
Will do. Firefox was telling me it was an unknown SSL cert issuer, if that helps.
Thanks.

Both you and vulture seem to be using firefox, so, I am wondering if that has something to do with it. Although,I am trying it on FF right now and haven't encountered the issue yet.So, looks like a bug that will be tough to pin down.

chel_of_the_sea
Posts: 27
Joined: Mon May 08, 2017 9:39 am

Re: Random thread

Post by chel_of_the_sea » Sat May 13, 2017 9:01 am

I only had the issue once. Haven't had it today.
:arrow: TL;DR: 27, trans woman, M.A. in math, Seattle area. Tutor by current trade, but in a bit of professional limbo (if you know anyone hiring, let me know!).

dsp
Posts: 6
Joined: Mon May 08, 2017 11:24 pm

Re: Random thread

Post by dsp » Sat May 13, 2017 7:04 pm

I have been using firefox and have experienced no problems.

ASol
Posts: 3
Joined: Mon May 15, 2017 3:31 am

Re: Random thread

Post by ASol » Mon May 15, 2017 3:42 am

I'm taking a stats/R course over the summer thru Coursera. "Statistics with R Specialization," offered by Duke University. I'm only a week in, so I have no idea how in depth the stats get. Apparently Gwern took the same course, and reviewed it positively.

Apparently a certificate is conferred at the end of the 5 courses in the module. Does anyone know if it's worth anything?

Raininginsanity
Posts: 51
Joined: Tue May 23, 2017 4:50 am

Re: Random thread

Post by Raininginsanity » Tue May 23, 2017 5:32 am

phaedrus wrote:
Mon May 08, 2017 5:53 pm
To start: what are good resources for learning statistics? I'm learning data science, but I want to know what exactly it is that I'm using scipy for.
What side of statistics is it that you're most interested in? The math? The programming? The applied side? For me the applied side is the most fun whereas the math is still essentially voodoo, and the programming Greek.

phaedrus
Posts: 21
Joined: Mon May 08, 2017 4:01 pm

Re: Random thread

Post by phaedrus » Sun May 28, 2017 1:00 am

Raininginsanity wrote:
Tue May 23, 2017 5:32 am
phaedrus wrote:
Mon May 08, 2017 5:53 pm
To start: what are good resources for learning statistics? I'm learning data science, but I want to know what exactly it is that I'm using scipy for.
What side of statistics is it that you're most interested in? The math? The programming? The applied side? For me the applied side is the most fun whereas the math is still essentially voodoo, and the programming Greek.
I can do the programming, but I'd like to not be cargo-culting the math. I can generate pretty pictures with regression testing and whatnot, but I don't know what any of it means, and I can feel Nassim Taleb giving me the evil eye all the way from a squid ink restaurant somewhere in Italy.

Raininginsanity
Posts: 51
Joined: Tue May 23, 2017 4:50 am

Re: Random thread

Post by Raininginsanity » Sun May 28, 2017 6:23 am

There's a few things that really helped me to understand statistics at a more intuitive level. Understanding probability and distributions is the first step. Everything else builds off of that. What I did after that was that I created a list of all the tests that I could think of (t-test, ANOVA, ANCOVA, regression, correlation, partial correlation, chi-square, etc.), and went one by one and studied them enough that I could write a basic description of each. At first my descriptions were very basic "t-tests compare two distributions", "ANOVAs compare 2 or more distributions", "ANCOVAS compare 2 or more distributions while controlling for a variable", etc. I didn't worry about all the various forms of t-tests, or whether anova should be a subset or regression or even about p values or effect sizes. I built up my definitions over time as I went on to understanding more and more. Once I had a broad grasp of the types of situations in which the tests were used, then I studied more in depth to understand how to interpret results, effect sizes, and the different type of t-tests or whatever. Just having that broad overview of the major types of tests is significant, and feels like considerable progression even if you don't understand the exact definition of a p-value (which *is* important to understand, most people don't get it which is a shame).

SolenoidEntity
Posts: 18
Joined: Tue May 09, 2017 3:42 am

Re: Random thread

Post by SolenoidEntity » Wed Jun 21, 2017 1:51 am

The Kansas City Royals just brought in a pitcher named Scott Alexander. I'm sure there are many people named that, but my first reaction was, "Wait, he uses his pseudonym when he pitches, too?"

Post Reply