« first day (1726 days earlier)      last day (3056 days later) » 

Meta
Meta
10:04
0
Q: Adding a new tag to the Mathematica forum

Michael McCainHow are new tags added to the Mathematica forum? Lately I've been working with celltags in Mathematica, and there have been quite a few questions on them in this forum. I think they are important enough to have their own tag in the forum. What is the procedure for requesting this? I am assu...

feature-request
 
3 hours later…
Szabolcs
Szabolcs
12:52
@halirutan Maybe @TimStone can help (I pinged him too)? Or a moderator could as SE directly? (@R.M., @J.M., @MrWizard)
 
2 hours later…
Michael E2
Michael E2
14:35
@sky-light W|A is using Mma V11. Or maybe you aren't using V9 effectively. Hard to tell without an example.
sky-light
sky-light
14:54
@MichaelE2 Hi
Please see this example
http://www.wolframalpha.com/input/?i=Solve%5Ba%2F(1+%2B+b+x)+-+Log%5B1+%2B+b+x%5D+%2B+1+%3D%3D+0,+x%5D
Michael E2
Michael E2
15:16
@sky-light See i.sstatic.net/PyNM8.png -- By "effectively," I meant there are many ways to solve equations in Mma. I believe W|A often sends multiple commands to different servers, and reports the results that come back within the timeout limit.
sky-light
sky-light
15:39
@MichaelE2 I agree with you. However, for someone beginner, it is not easy to find these ways. Thank you.
Michael E2
Michael E2
@sky-light No, they're not all easy to find. :)
kirma
kirma
16:21
How to model stock index data like Dow Jones or S&P? Apparently HMM on Gaussians isn't really enough to reproduce stock market crashes.
 
3 hours later…
blochwave
blochwave
19:29
@kirma that's the million-dollar question, right? :-D
Anyone who knows ain't telling...
 
3 hours later…
Searke
Searke
22:38
@kirma I had trouble getting into financial math when I started because I asked the same question.
Everything I read always began with that assumption, which is testable:
GEDelta = Differences@FinancialData["GE", "Jan. 1, 2000"][[All, 2]];
DistributionFitTest@GEDelta
0
If you haven't read the work of Nassim Taleb, I'd recommend it.
Searke
Searke
23:09
@kirma the differences are closer to being Cauchy distributed than normally distributed. Unfortunately, when assume that the deltas in prices are Cauchy distributed instead of normally distributed, so much financial math stop making sense.
For example, the logic of portfolio diversification goes out the door. You can't reduce the variance of your portfolio by distributing your risk over different uncorrelated companies.

« first day (1726 days earlier)      last day (3056 days later) »