What technique would you recommend everyone learns?

TheEcologist

Global Moderator
#1
Hi Everyone,

I was wondering, which specific knowledge/ technique you think is the most useful thing you have learned in the realm of statistics and math?

For me it's definitely when I first started understanding and using "maximum likelihood approaches", what an eye-opener! A world of previously unfathomed opportunities and insights just opened up for me.

Before that, it was R, as a concept and language that got me started on this road back in 2004 (within R, the most useful path I took was the path of the "R purist", also known as the "dark side", a wrapper free Spartan approach, the awesome graphics creating road, the straight and narrow ect ect - but that is a story for a different time ;) )

So what about you guys? And can you convince us to learn this?

Looking forward to your responses.

M.
 

Rhodo

New Member
#2
I really like this idea for a topic. Perhaps people could share a good resource to learn more about their suggestions as well.
 

vinux

Dark Knight
#3
There are many for me. I was more interested in mathematics till I started working in corporate.

Mathematics
I enjoy most part of mathematics like topology, graph theory, mathematical analysis ... etc. I think the following is useful in general
(1) Logic/Mathematical proofs. For fun read this
(2) Geometry. This would simplify your life. When I learned polar coordinate, it was like a big opportunity to explore my graphics experiments. I have created lot of screen savers in OpenGL.

Statistics
I have more to add on each items.
(1) glm, GAM, bootstrap,
(2) Bayesian analysis
(3) I didn't like R initially, I used S+ till 2005. I started using R again in 2010. I agree with Ecologist on the simplicity aspect.

Other Stuff
These are not really connected to math/stat learning.
(1) Latex
(2) C/Python
 

bugman

Super Moderator
#4
I am still crap at math. But for me it was - R, mixed models (thanks Jake for steering me in the right direction), GAM, GLM's, the fundmentals of time series, PERMANOVA, distance measures (thanks Legendre and Lengendre) and that Raptors will destroy all of humanity if there little alligator hands will let them...

(sorry for the last bit).
 

vinux

Dark Knight
#7
Excuse my ignorance, but in a nut shell,. what is graph theory? I'm intrigued.
Vinux still needs to answer this question, imo. :)
Thank you TE;sorry bugman. :eek:I am not sure how I missed this question.

Graph theory is the study of graphs (not really about graphic). In R context, the graph is produced using igraph package. You get well articulated information about graph theory in wikipage.

There are some interesting problems can be easily solved using graph theory.
Classic example is Seven Bridges of Konigsberg.
Other problems are map colouring, travelling sales man problem, marriage problem, etc.