Karol Piczak

999 reputation

31022

bio	website	karol.piczak.com
	location	Warsaw, Poland
	age	26
visits	member for	2 years, 3 months
	seen	2 days ago
stats	profile views	206

One-liner: Mashup of Finance, IT/CS/AI/ML and other buzz words. Garnish with music composition and some sound design.

Educational background: Master of Finance, now pursuing a PhD in Computer Science (audio data analysis in recommender systems).

My online ID: karol.dvl.pl

	999 reputation	bio	website	karol.piczak.com	visits	member for	2 years, 3 months
31022 badges		location	Warsaw, Poland		seen	2 days ago

summary answers questions tags badges favorites bounties reputation activity

107 Actions

suggestions reviews revisions comments badges posts accepts all

Mar 27	awarded	Nice Question
Mar 20	comment	Keeping a track record honest You could timestamp only the article content (like: "all above this line is signed and trusted not to be modified since it's creation at ..."), put the stamp in a separate field or even use timeMarker's Notarize Link (example for this question: bit.ly/eRxVQ6). Though in the end it won't matter much how you do it. Simply showing your commitment to credibility should be sufficient for most readers (I doubt they are gonna check it either way). And those who don't want to trust you will still find a plentiful supply of reasons supporting their point.
Mar 20	revised	Keeping a track record honest Fix minor stuff
Mar 20	answered	Keeping a track record honest
Mar 17	comment	Switching from C++ to R - limitations/applications I think I will give it a try then. Thanks. It's nice you can count on getting answers from experts here no matter what the topic. :-)
Mar 17	accepted	Switching from C++ to R - limitations/applications
Mar 17	answered	Bank of England base rate feed
Mar 16	awarded	Editor
Mar 16	revised	Are public historical time series available for ratings of sovereign debt? Added Moody's and S&P links
Mar 16	answered	Are public historical time series available for ratings of sovereign debt?
Mar 15	comment	Switching from C++ to R - limitations/applications Yeah, I suppose so. But this part "at a possible cost in terms of time to code" is exactly my problem. I indeed like C++, but at times it's really hard to get something up and running quickly. So I'm looking for a tool that would allow me to crash test my concepts more rapidly and see if they are worth implementing in, as you name it, industrial-strength at all. But I wouldn't like to shoot myself in the foot either and get locked in a point where it's not appropriate for production use nor preliminary research.
Mar 15	comment	Switching from C++ to R - limitations/applications Thanks for bringing this up. I haven't thought that memory management could be such an issue. And great example selection. Just what gets into my field of interests. ;-) At least I shall know what not to expect from R. But as you say, it's probably nice to at least get acquainted with it, so as to know when it may come in handy. Forgive my question blatantly arising from my personal interests and lack of knowledge in the matter at the same time, but how appropriate is R in machine learning applications and particularly Bayesian network inference?
Mar 15	comment	Methods for pricing options +1 Great answer. I like how it organizes the knowledge.
Mar 15	comment	Switching from C++ to R - limitations/applications I suppose this would fit SO (or maybe stats.SE?) as well, but I assumed new questions are of more use in here. I hope it's sufficiently on topic here though.
Mar 15	asked	Switching from C++ to R - limitations/applications
Mar 10	awarded	Teacher
Feb 3	awarded	Suffrage
Feb 1	awarded	Scholar
Feb 1	accepted	What methods do you use to improve expected return estimates when constructing a portfolio in a mean-variance framework?
Feb 1	comment	What methods do you use to improve expected return estimates when constructing a portfolio in a mean-variance framework? Thanks for a detailed answer. Haven't heard of Michaud's approach at all, so I will have to get into this paper in some spare time. And indeed - good return estimates are a value in themselves. Heck, I wouldn't need fancy theories when I could absolutely trust my return forecasts. ;-)