Simra's Mega-Feed

Reading list: Predicting stock prices with HMMs

Mon, 03 Jun 2013 05:04:21 +0000

I stumbled on the scikit-learn HMM toolkit and thought I'd share some reading materials on HMMs. I personally dislike models that require a lot of hyperparameters (or a least some set of hyperparemeters that are hard to guess a priori, like the number of states), but that said I like the HMM's ability to abstract the state of a random process.
I would love to play with these in other contexts too...
Most of these papers came from a Bing search:

Proof that most methods of making stock market predictions don’t work

Sun, 19 May 2013 04:13:07 +0000

[5/19: fixed the pdf link]
Summary of a Vanguard study:
http://monevator.com/methods-of-making-stock-market-predictions-dont-work/
Link to the study:
https://personal.vanguard.com/pdf/s338.pdf

Note that the methodology is applied only to predicting whole market returns: it isn't applied to choosing individual stocks.

Money quote: "... most of the variables they studied were about as helpful as asking your dog for one bark or two to tell you whether the market was headed up or down."

And some discussion about application of the approach to individual stocks here
http://www.bogleheads.org/forum/viewtopic.php?t=66143

MtGox DDOS

Sun, 21 Apr 2013 19:25:57 +0000

Bitcoin exchange MtGox is unavailable due to DDOS attack. https://m.facebook.com/MtGox/posts/459849287432817 Bitcoin value plunged by 20% on btc-e as a result. This is a crazy volatile horse trading market. My sense is that arbitrage opportunities are not only limited by liquidity but also by daily transaction limits.

simra.net returns

Robert Sim — Sat, 20 Apr 2013 08:35:00 +0000

This blog has been idle since 2008. In the meantime I've had simra.net up and down a few times but never any really active posting. Now I have the site back up and running, with a twist: the focus of the site is algorithmic trading, HFT, Bitcoin arbitrage, etc.

How to really make money with Bitcoin

Sat, 20 Apr 2013 02:02:38 +0000

Bitcoin is the new speculation hotness. Never mind that it costs more in energy to create a bitcoin than the coin is actually worth.

Recently a friend pointed me to some code facilitating arbitrage between bitcoin markets. This sounds interesting, though I imagine it is somewhat difficult to get a really profitable edge- the barrier to entry seems very low. On the other hand, someone else has found a surefire way to profit from the bitcoin speculation bubble: http://boingboing.net/2013/04/18/which-makes-more-money-mining.html

How are we doing so far?

Fri, 05 Apr 2013 05:31:26 +0000

Various tweaks to the model over the last few days, as well as some backdating and thinking about how to make picks. I'm still working on some formatting/html stuff to make this stuff more browsable but in general you can find the picks after the close each day at, for example, /stocks/topStockPicks-2013-04-03.html, and the latest at /stocks/latestStockPicks. Full backdated index at http://simra.net/stocks.

I'm running the job by hand each night, for now. Small cap equities have largely been advancing lately which makes beating the market harder. For now I'm looking for thresholds so I can have better-than-random precision at picking stocks that advance for the day. This is not the same as ROI but we'll get there. Currently, score thresholds above about 50 are looking promising, though they are relatively rare. Suffice to say this is a hard problem space. :-)

Graph: precision/recall curve for 03-14 to 03-28. The red curve would be random guessing.

Stock picks for 3/29

Sat, 30 Mar 2013 06:19:26 +0000

First trial run of the model based on today's closing prices. Let's see how they look next Friday, shall we?
http://simra.net/stocks/latestStockPicks.html

Rank Ticker
1 CECO
2 GTAT
3 TTI
4 DM
5 CWTR
6 IQNT
7 LAWS
8 COCO
9 MOV
10 NVTL
11 SFY
12 LINC
13 DGII
14 WGO
15 SAH
16 EBIX
17 CBR
18 ENTR
19 STRI
20 AKS

The emoticons of pi

Fri, 15 Mar 2013 06:50:56 +0000

Happy Math.Pi day!
3.
1415926535 8979323846 2643383279 5028841971 6939937510 5820974944 5923078164 0628620899 8628034825 3421170679 8214808651 3282306647 0938446095 5058223172 5359(Q28 4811174502 8410270193 8521105559 6446229489 5493038196 4428810975 6659334461 2847564823 3786783165 2712019091 4564856692 3460348610 4543266482 1339360726 0249141273 7245870066 0631558817 4881520920 9628292540 9171536436 7892590360 0113305305 4882046652 1384146951 9415116094 3305727036 5759591953 0921861173 8193261179 3105118548 0744623799 6274956735 1885752724 8912279381 8301194912 9833673362 4406566430 8602139494 6395224737 1907021798 6094370277 0539217176 2931767523 8467481846 7669405132 0005681271 4526356082 7785771342 7577896091 7363717872 1468440901 2249534301 4654958537 1050792279 6892:\35 4201995611 2129021960 8640344181 5981362977 4771309960 5187072113 4999999837 2978049951 0597317328 1609631859 5024459455 3469083026 4252230825 3344685035 2=]11881 7101000313 7838752886 5875332083 8142061717 7669147303 5982534904 2875546873 1159562863 8823537875 9375195778 1857780532 1712268066 1300192787 6611195909 2164201989 3809525720 1065485863 2788659361 5338182796 8230301952 0353018529 6899577362 2599413891 2497217752 8347913151 5574857242 4541506959 5082953311 D=727855 8890750983 8175463746 4939319255 0604009277 0167113900 9848824012 8:$16035 6370766010 4710181942 9555961989 4676783744 9448255379 7747268471 0404753464 6208046684 2590694912 9331367702 8989152104 7521620569 6602405803 8150193511 2533824300 35:L4024 7496473263 9141992726 0426992279 6782354781 6360093417 2164121992 4586315030 2861829745 5570674983 8505494588 5869269956 9092721079 7509302955 3211653449 8720275596 0236480665 4991198818 3479775356 6369807426 5425278625 5181841757 4672890977 7727938000 8164706001 6145249192 1732172147 7235014144 1973568548 1613611573 5255213347 5741849468 4385233239 0739414333 4547762416 8625189835 6948556209 9219222184 2725502542 5688767179 0494601653 4668049886 2723279178 6085784383 8279679766 8145410095 3883786360 9506800642 2512520511 7392984896 0841284886 2694560424 1965285022 2106611863 0674427862 2039194945 0471237137 8696095636 4371917287 4677646575 7396241389 0865832645 9958133904 7802759009 9465764078 9512694683 9835259570 9825822620 5224894077 2671947826 8482601476 9909026401 3639443745 5305068203 4962524517 4939965143 1429809190 6592509372 216964=3 57098:&7 4105978859 5977297549 8930161753 9284681382 6868386894 2774155991 8559252459 5395943104 9972524680 8459872736 4469584865 3836736222 6260991246 0805124388 4390451244 1365497627 8079771569 1435997700 1296160894 4169486855 5848406353 4220722258 2848864815 8456028506 0168427394 5226746767 8895252138 5225499546 6672782398 6456596116 3548862305 7745649803 5593634568 1743241125 1507606947 9451096596 0940252288 7971089314 5669136867 2287489405 6010150330 8617928680 9208747609 1782493858 9009714909 6759852613 6554978189 3129784821 6829989487 226:P485 7564014270 4775551323 7964145152 3746234364 5428584447 9526586782 1051141354 7357395231 1342716610 2135969536 2314429524 8493718711 0145765403 5902799344 0374200731 0578539062 1983874478 0847848968 3321445713 8687519435 0643021845 3191048481 0053706146 8067491927 8191197939 9520=)96 6342875444 0643745123 7181921799 9839101591 9561814675 1426912397 4894090718 6494231961
5679452080 9514655022 5231603881 9301420937 6213785595 6638937787 0830390697 9207734672 2182562599 6615014215 0306803844 7734549202 6054146659 2520149744 2850732518 6660021324 3408819071 0486331734 6496514539 057962D8 1005508106 6:O69981 6357473638 4052571459 1028970641 4011097120 6280439039 7595156771 5770042033 7869936007 2305:L31 7635942187 3125147120 5329281918 2618612586 7321579198 4148488291 6447060957 5270695722 0917567116 7229109816 9091528017 3506712748 5832228718 3520935396 5725121083 5791513698 8209144421 0067510334 6711031412 6711136990 86:36398 3150197016 5151168517 1437657618 3515565088 4909989859 9823873455 2833163550 7647918535 8932261854 8963213293 3089857064 2046752590 7091548141 6549859461 6371802709 8199430992 4488957571 2828905923 2332609729 9712084433 5732654893 8239119325 9746366730 :$041428 1388303203 8249037589 8524374417 0291327656 1809377344 4030707469 2112019130 2033038019 7621101100 4492932151 6084244485 9637669838 9522868478 3123552658 2131449576 8572624334 4189303968 6426243410 7732269780 2807318915 4411010446 8232527162 0105265227 2111660396 6655730925 4711055785 3763466820 6531098965 2691862056 4769312570 5863566201 8558100729 3606598764 8611791045 3348850346 1136576867 5324944166 8039626579 7877185560 8455296541 2665408530 6143444318 5867697514 5661406800 7002378776 5913440171 2749470420 5622305389 9456131407 1127000407 8547332699 3908145466 464:P797 2708266830 6343285878 5698305235 8089330657 5740679545 7163775254 2021149557 6158140025 0126228594 1302164715 5097925923 0990796547 3761255176 5675135751 7829666454 7791745011 2996148903 0463994713 2962107340 4375189573 5961458901 9389713111 7904297828 5647503203 1986915140 2870808599 0480109412 1472213179 4764777262 2414254854 5403321571 8530614228 8137585043 0633217518 2979866223 7172159160 7716692547 4873898665 4949450114 6540628433 6639379003 9769265672 1463853067 3609657120 9180763832 7166416274 8XP07869 2560290228 4721040317 2118608204 1900042296 6171196377 9213375751 1495950156 6049631862 9472654736 4252308177 0367515906 7350235072 8354056704 0386743513 6222247715 8915049530 9844489333 0963408780 76932;]9 7805419341 4473774418 4263129860 80998XD7 4132604721 5695162396 :@573021 6315981931 9516735381 2974167729 4786724229 2465436680 0980676928 2382806899 6400482435 4037014163 1496589794 0924323789 6907069779 4223625082 2168895738 3798623001 5937764716 5122893578 6015881617 5578297352 3344604281 5126272037 3431465319 7777416031 9906655418 7639792933 4419521541 3418994854 4473456738 3162499341 9131814809 2777710386 3877343177 2075456545 3220777092 1201905166 0962804909 2636019759 8828161332 3166636528 =]266863 3606273567 6303544776 2803504507 7723554710 5859548702 7908143562 4014517180 6246436267 9456127531 8134078330 3362542327 8394497538 243720:# 3114771199 2606381334 6776879695 9703098339 1307710987 0408591337 4641442822 7726346594 7047458784 7787201927 7152807317 6790770715 7213444730 6057007334 9243693113 8350493163 1284042512 1925651798 0694113528 0131470130 4781643788 5185290928 5452011658 3934196562 1349143415 9562586586 5570552690 4965209858 0338507224 2648293972 8:/83163 0577775606 8887644624 8246857926 0395352773 4803048029 00:L0758 2510474709 1643961362 6760449256 2742042083 2085661190 6254543372 1315359584 5068772460
2901618766 7952406163 4252257719 5429162991 9306455377 9914037340 4328752628 8896399587 9475729174 6426357455 2540790914 5135711136 9410911939 3251910760 2082520261 8798531887 70:*9725 9167781314 9699009019 2116971737 2784768472 6860849003 3770242429 1651300500 5168323364 3503895170 2989392233 4517220138 1280696501 1784408745 1960121228 ;]716231 3017114448 4640903890 6449544400 6198690754 8516026327 5052983491 8740786680 8818338510 2283345085 0486082503 9302133219 7155184306 3545500766 8282949304 1377655279 3975175461 3953984683 3936383047 4611996653 8581538420 5685338621 8672523340 2830871123 2827892125 0771262946 3229563989 8989358211 6745627010 2183564622 0134967151 8819097303 8119800497 3407239610 3685406643 1939509790 1906996395 5245300545 0580685501 9567302292 1913933918 5680344903 9820595510 0226353536 1920419947 4553859381 0234395544 9597783779 0237421617 2711172364 3435439478 2218185286 2408514006 6604433258 8856986705 4315470696 5747458550 3323233421 0730154594 0516553790 6866273337 99:31562 5784322988 2737231989 8757141595 7811196358 3300594087 3068121602 8764962867 4460477464 9159950549 7374256269 0104903778 1986835938 1465741268 0492564879 8556145372 3478673303 9046883834 3634655379 4986419270 5638729317 4872332083 7601123029 9113679386 2708943879 9362016295 1541337142 4892830722 0126901475 4668476535 7616477379 4675200490 7571555278 1965362132 3926406160 1363581559 0742202020 3187277605 2772190055 6148425551 8792530343 5139844253 2234157623 3610642506 3904975008 6562710953 5919465897 5141310348 2276930624 7435363256 9160781547 QQ152843 6679570611 0861533150 4452127473 9245449454 2368288606 1340841486 3776700961 2071512491 4043027253 8607648236 3414334623 5189757664 5216413767 9690314950 1910857598 4423919862 9164219399 4907236234 6468441173 9403265918 4044378051 3338945257 4239950829 6591228508 5558215725 0310712570 1266830240 2929525220 1187267675 6220415420 5161841634 8475651699 9811=)01 0029960783 8690929160 3028840026 9104140792 8862150784 2451670908 7000699282 1206604183 7180653556 7252532567 5328612910 4248776182 5829765157 9598470356 2226293486 0034158722 9805349896 5022629174 8788202734 2092222453 3985626476 6914905562 8425039127 5771028402 7998066365 8254889264 8802545661 0172967026 6407655904 2909945681 5065265305 3718294127 0336931378 5178609040 7086671149 6558343434 7693385781 7113864558 7367812301 4:L87126 6034891390 9562009939 3610310291 6161528813 8437909904 2317473363 9480457593 1493140529 7634757481 1935670911 0137751721 0080315590 2485309066 9203767192 2033229094 3346768514 2214477379 3937517034 4366199104 0337511173 5471918550 4644902636 5512816228 8244625759 1633303910 7225383742 1821408835 0865739177 1509682887 4782656995 9957449066 1758344137 5223970968 3408005355 9849175417 3818839994 4697486762 6551658276 5848358845 3142775687 9002909517 0283529716 3445621296 4043523117 6006651012 4120065975 :3276178 :&292041 9748442360 8007193045 7618932349 2292796501 9875187212 7267507981 2554709589 0455635792 1221033346 6974992356 3025494780 2490114195 2123828153 0911407907 3860251522 7429958180 7247162591 6685451333 1239480494 7079119153 2673430282 4418604142 6363954800 0448002670 4962482017 9289647669 7583183271 3142517029 6923488962 7668440323 2609275249 6035799646 9256504936 8183609003 2380929345
9588970695 3653494060 3402166544 3755890045 6328822505 4525564056 4482465151 8754711962 1844396582 5337543885 6909411303 1509526179 3780029741 2076651479 3942590298 9695946995 5657612186 5619673378 6236256125 2163208628 6922210327 4889218654 3648022967 8070576561 5144632046 9279068212 0738837781 4233562823 6089632080 6822246801 2248261177 1858963814 0918390367 3672220888 3215137556 0037279839 4004152970 0287830766 7094447456 013458)7 2543709069 7939612257 1429894671 5435784687 8861444581 2314593571 9849225284 7160504922 1242470141 2147805734 5510500801 9086996033 0276347870 8108175450 1193071412 2339086639 3833952942 5786905076 4310063835 1983438934 1596131854 3475464955 6978103829 3097164651 4384070070 7360411237 3599843452 2516105070 2705623526 6012764848 3084076118 3013052793 2054274628 6540360367 4532865105 7065874882 2569815793 6789766974 2205750596 8344086973 5020141020 6723585020 0724522563 2651341055 9240190274 2162484391 4035998953 5394590944 0704691209 1409387001 2645600162 3742880210 9276457931 0657922955 2498872758 4610126483 6999892256 9;D81592 0560010165 5256375678

digits of pi: http://www.joyofpi.com/pi.html
Emoticons: http://en.wikipedia.org/wiki/List_of_emoticons

Simple conversion of the Wikipedia emoticons to numeric representation and then lookup in PI. I might have missed a few because in order to preserve formatting I kept the whitespace...

Vatsal Shah survey of methods for price prediction

Fri, 08 Mar 2013 01:12:51 +0000

Interesting research paper. Once I've read it completely I'll share some thoughts.
http://www.vatsals.com/Essays/MachineLearningTechniquesforStockPredictio...

UPDATE 3/16: Ok, I read this over and there's not really anything there of interest. As an aside I ran some experiments last week but had to rebuild my system shortly after due to a HD failure. Once I get the data back from my backup I'll share.

John Langford on Stock Prediction with ML

Fri, 08 Mar 2013 01:03:39 +0000

Fellow MSFTie John Langford shares his thoughts on stock prediction. http://hunch.net/?p=210

A first foray into predicting stock prices

Tue, 05 Mar 2013 13:47:44 +0000

Pasting below some old correspondence with a friend. This was my first cut at applying a machine learning approach to trading.

Thought I'd share the results I had from modeling stock price jumps. I took another stab at it this past week, here's how I did it:
I collected historical prices from Yahoo for all SP500 tickers. Training data comes from every trading day of 2011, test data from the first 6 months of 2012.
Features: for a given ticker, take two weeks of closing prices and compute the percent change for each day, relative to the first closing price, and also relative to the last closing price. I also threw in the same percent change for SPY as a market indicator. Tried throwing in VIX, both relative and absolute values but it made things worse.

So that gives about 60 features (15 x 2 for the ticker and another 15x2 for SPY) for each ticker, and we have a sample for every trading day in 2011 after the first two weeks, and for every ticker in the S&P500. == ~100,000 training examples.

Then I labeled the examples by looking at the closing price five trading days after the last close, and marking examples where there was more than 5% gain as positives. Everything else was marked as a negative.

Then I trained a binary classifier on the training examples. I used logitboost with 1000 stumps but there's a lot of alternative options you can try.

Finally, score the test samples from 2012 and compute a ROC curve (for each possible score threshold, how many examples exceeding the threshold are actually positives (true positives) and how many are actually negatives (false positives). See wikipedia for details.

First attached image is the ROC plot. Blue curve is the classifier performance and red is random guessing performance (in the absence of a prior- which I'll come to). TP Rate= true positives/All positives FP rate=false positives/All negatives.

So this is an encouraging first step- we do better than random guessing. The problem is that the prior is not 50-50. In the test set only about 7% of the examples saw a 5% increase, so it's not enough to 'just' perform better than random. We want to know the precision, which is #True positives/(#True positives+#false positives), which depends on the prior. The second attached image is the Precision vs the Recall. (Recall is just another name for TP rate). So if we were to put this classifier out there making calls on which stocks to buy, it would only be right about 10-15% of the time and miss more than half the opportunities (eg if we chose a threshold in the 30% recall range we'd have about 12% precision).

Ok, then, I thought maybe 5% bumps are too rare and I should just try to build a classifier that predicts pos vs negative price swings. So I relabeled the examples where a gain over the next week was a positive and a loss was a negative. Third attached image is the result, which shows that we would do a little bit worse than random guessing... :-(

nobody@flickr.com (simra) — Sun, 19 Jul 2009 06:37:01 +0000

simra posted a photo:

Vancouver Folk Music Festival.

I'm a little disappointed because the color settings were out of whack on the camera, but it's still a great shot of the audience, as well as the action on Stage 4. We caught the tail end of a high-intensity set- I wish I'd been there for more.

Vancouver Folk Music Festival

nobody@flickr.com (simra) — Sun, 19 Jul 2009 06:28:31 +0000

Robert Sim — Wed, 26 Nov 2008 23:43:00 +0000

Testing 123

Robert Sim — Tue, 18 Nov 2008 22:16:00 +0000

I'm experimenting with merging my flickr feed, google shared items, and this blog feed all under one super-feed hosted by Yahoo pipes. There may be some repeat posts, so bear with me. Everything will still be piped through feedburner so you shouldn't have to make any adjustments...

ChChChChanges

Robert Sim — Mon, 17 Nov 2008 20:38:00 +0000

1. I'm losing my hosting provider. :-( In the meantime we'll camp out over here at blogger. Once the dust settles on #2 and #3 I'll look into reviving all the old posts.

2. Where'd ya go? I was looking for a new job. It takes time to find a research job that's a good fit, so hence the longish blackout. There's so much of me that's already on the net already but I figured I'd err on the side of caution. Hopefully my new employer won't be shocked by my endless irreverence. Which brings us to...

3. I've got a new job! The story is long, complicated, and unbelievably dull, but I'm heading back to Microsoft to join the Windows Live Experience Team in a research capacity. No robots or vision for a while... I look at it as the same toolbox-- new problems. It's very exciting to think that I am getting into an area of machine learning research that could impact literally millions of people. Hopefully for the better... :-)

So long, Braintech, and thanks for all the fish. We did some great stuff over there but in the end it was not meant to be. If you're a vision scientist looking for work, feel free to contact me and I can put you in touch with them. simra at ubc cs still works, so does robert.sim 's gmail.

Oh crap he's back.

Robert Sim — Sat, 15 Nov 2008 00:55:00 +0000

More to come... :-)

I've moved (again)

Robert Sim — Tue, 30 May 2006 10:56:00 +0000

Update May, 2007: My new blog is simra.net/blog.
Previous post...
Please visit my (old) new typo blog at simra.net.

Robot hazing

Robert Sim — Sat, 15 Apr 2006 10:22:00 +0000

Title says it all

Land disputes in BC

Robert Sim — Tue, 28 Feb 2006 10:27:00 +0000

Long time no see...

The Tyee has an eye-opening article this morning concerning public access rights and the sad legacy of land entitlement in BC.

music of the week

Robert Sim — Sun, 29 Jan 2006 23:15:00 +0000

The Friday morning waitress at the Naam has great musical taste and for a while she had a fantastic mix tape of Iron&Wine, the Velvet Underground and a handful of other great tunes. When it slipped off the playlist in favour of a new album I was a bit disgruntled, but the replacement is absolutely amazing- Sufjan Stevens' (C'mon feel the) Illinoise is the kind of music that wedges itself in your brain and won't let go. My first impression was that he's got Paul Simon's voice, but the album sounds more like classic Mahavishnu Orchestra mixed with Stereolab mixed with The Polyphonic Spree (my first exposure to which was on Austin City Limits). Everything old is new again...

No links tonight... google them on your own. !!

Election post-mortem

Robert Sim — Tue, 24 Jan 2006 09:44:00 +0000

No big surprises last night, but it was satisfying to see Sam Bulte lose her seat. The Vancouver Sun's headlines this morning made big on the fact that the PC's were shut out of every seat in Toronto, Vancouver and Montreal. While the West definitely decided this election (proving Ontario is not the stronghold the Reformers used to gripe about), the split is still very regional- rural/urban, Atlantic/Quebec/Ontario/Praries/BC. Here's hoping a Liberal/NDP block can keep the more radical PC's in check.

January in Vancouver

Robert Sim — Fri, 06 Jan 2006 09:40:00 +0000

It's well over 10 degrees C here this morning, the grass is nice and green and there's even some blue sky peeping through the clouds. Great reasons to love living in this city... :-)

Election coverage...

Robert Sim — Wed, 04 Jan 2006 10:30:00 +0000

Last post for today, honest. As a follow-up to the canadian election, The Tyee, which, in my opinion, is the best independent source of news coverage in Canada, has re-opened its election superblog. Follow the blow by blow...

Happy new year, partie le trois..

Robert Sim — Wed, 04 Jan 2006 10:10:00 +0000

Miscellanea: We're just back from our annual holiday visit to hogtown to see family, old friends, etc. The big news is Ravi Bhat successfully defended his PhD in late Dec, which was an extra reason to crack open the bubbly. We also had a great trip to Kingston and Montreal before flying back to Vancouver. Our friends Kelly and Matt (and kids and pets, etc) have a wonderful place in the woods north of K-town and our visit there is always the most relaxing thing I've done all year.

Post-holiday playlist:

Phish -- Round Room
Grateful Dead -- Workingman's Dead
Coldplay -- Parachutes
The Beatles -- Rubber Soul
Sarah Harmer -- I'm a Mountain

Now, on with the show.

Happy new year, part the deux.

Robert Sim — Wed, 04 Jan 2006 09:51:00 +0000

It's election time in Canada and Boingboing has been running a series of posts on incumbent Liberal MP Sam Bulte's relationship with the entertainment industry. The main thrusts of the story are that Bulte is the industry's champion for copyright reform in favour of clamping down on user's rights, and furthermore, Bulte is 'on the take' so to speak, receiving campaign financing from the industry and benefitting from industry-sponsored campaign fundraisers. Bulte seems to have forgotten that her role as an MP is to represent her constituents, rather than whoever's lining her pockets.

While Bulte is exhibiting a certain style of American politics that doesn't wash well here north of the border, I'm very curious to know how many other Liberal and Conservative candidates have similar kinds of mutual back-scratching ties- to, say, pharmaceuticals, forestry, agriculture or mining. My sense is that Bulte is only the very public tip of a deep, ugly iceberg.

Happy new year, part un.

Robert Sim — Wed, 04 Jan 2006 09:36:00 +0000

A slurry of posts to ring in the new year. First off, this month's Wired magazine has a good, if fluffy, article on Stanley, the robotic vehicle that won the Grand Challenge a few months back.

Sebastian Thrun gave a talk at NIPS last month on their success, and the general muttering that I heard from the crowd was that it was thin on any brilliant technical breakthroughs. People seem to miss the fact that Stanley was one of only a handful of vehicles to actually finish the race. Beyond that, the deeper message, and one that I think is important for robotics research, is that success with real-time systems is not so much about engineering the system to death but coming up with simple, fast hacks that get the job done. This was our experience in the AAAI competitions as well. The down-side of this approach is that you wind up with brittle systems that work well in the desert, say, but break in the city (and vice versa).

As an aside, I had a chance to take a ride in Stanley in Nov 2004, shortly after the team finished their first desert trial. Sebastian and Mike Montemerlo were ecstatic that they had completed roughly 7 miles of the 2004 course. They were looking like serious contenders right from the start. I asked Sebastian @ NIPS what his plans are for the future, and he sounds serious about city driving and a run from San Francisco to L.A.

Go green!

Robert Sim — Mon, 12 Dec 2005 19:10:00 +0000

If you're Canadian, take a few minutes to sign this petition to include the green party in the upcoming electoral debates: http://www.info-greenparty.ca/petition/.

More about it here.

Classic breakfasts and NIPS

Robert Sim — Fri, 02 Dec 2005 09:41:00 +0000

Friday is my favourite day of the week, not least because I manage to sneak out of the house early and head down to the Naam for breakfast. The Naam is special for so many good reasons- Kitsilano/Point Grey/UBC suffer from a serious lack of 24 hour restaurants, and the Naam offers something close to the Montreal breakfast I was spoiled on for so many years. Then there's the cozy fireplace and the usual trickle of people that make Kitsilano a great neighbourhood.

Back to work, though. This week we've got a series of pre-NIPS talks as people float into town for next week's big event. Yesterday Nic Schraudolph gave a nice talk on fast optimization methods, and later today we've got Aaron Hertzmann who does nice work in machine learning and graphics. ... and I'm also looking forward to Vinod Nair's talk on Monday, as his work looks to be related to my Ph.D. work. (Vinod is one of Geoff Hinton's students at UofT). Finally, there's NIPS itself, which promises to be as interesting as ever.

What's playing...

Robert Sim — Tue, 29 Nov 2005 14:25:00 +0000

Vapid content alert... in heavy rotation on my ipod:

Sarah Harmer -- I'm a mountain
Harry Manx -- Mantras for madmen
Harry Manx -- West eats meet
Iron & Wine -- Our endless numbered days
Indigo Girls -- All that we let in

... the thing about blogs.

Robert Sim — Tue, 29 Nov 2005 11:08:00 +0000

Well, there you have it. How many weeks since my last post? Not an auspicious start. This morning, it's snowing and slushy in Vancouver (a somewhat rare occurence), which makes me nostalgic for the east coast.

Just off a visit to Nick Roy @ CSAIL/MIT where I gave a talk about my work. We had a good week of work and managed to get out to a Blue Rodeo show, which, despite professing only a passing interest in the band, made the whole trip worthwhile. :-)

Why am I back? That was the question I left unanswered the other week. This blog might provide a clue: Bowen Island Journal. Bowen is a tiny island off the end of west van. We almost rented a place there when we moved here, but figured the 1.5hr commute wouldn't be all that much fun. Chris Corrigan's B.I.J. is one weblog that I return to again and again. He's managed to capture the essence of a place and its community so well. ... and it provides the kind of inspiration I'm looking for in articulating how blogging can really make a difference to the web. Chris' blog is pure content. There are no vapid "Check this out. LOL" links. ... and so, this is my goal for this blog- to capture a time and place, with a minimum of useless fluff. Not so much the physical place of Vancouver, or Canada, but life as an academic, as a scientist, a roboticist, a dad, a husband, etc. Over time, I hope this blog will produce a deeper portrait of life as I see it, or at least as I'm living it.

Wish me luck!

Preamble: how on earth did we get here?

Robert Sim — Thu, 03 Nov 2005 10:50:00 +0000

Some time in 1999 Scott Burlington and I wrote a little perl script called headline that screen-scraped stories off our favourite web sites, and posted them to a web page. At the time, there was no such thing as RSS, so we had to write custom code to scrape each site that we wanted headlines from. We ran the simple gauntlet of slashdot, wired, APOD, stock tickers, etc, and also posted links to the now-defunct Suck, which IMHO is still the uncontested all-time champ of web-based political commentary. It was great fun and I felt like we were playing a part in an amazing cultural renaissance. Of course, the main purpose for me was to entertain myself while procrastinating on my Ph.D. work (much as I'm doing right now- procrastinating on looking for faculty jobs. ;-)

Some time later I started thinking about how I could make my web site a little more dynamic and crafted together a rudimentary weblog. Of course, I didn't call it a blog at the time, and I'd never heard of Movable Type, but once it went live, I was hooked. I used the blog to post links I found entertaining, original book reviews, movie reviews, and then later it became a political tool for informing the masses about McGill's IP policy and other doings in my role as Grad Senator and later VP Academic for the grad society.

But there's a problem with blogging, and web culture in general and it's this: like the space occupied by electrons orbiting an atom, the web is mostly vacuous- devoid of content. We've created a device in which millions of monkeys type away (taking time away from other productive activites) all to produce commentary. The few monkeys who actually contribute content to the net act as giant lint brushes, ever collecting the detritus of links from the blogosphere. What are trackbacks other than a kind of showmanship- my lint collection is bigger than yours! Na-na-nana-naaa...

Now I must admit here, that I played my own small, embarassing part in this collective alternation between navel-gazing and hero worship. I lost countless hours scouring metafilter, fark and boingboing, looking for whatever it is that would give me a rush of pleasure and/or self-righteous indignation for the morning and make me look forward to eventually getting down to work.

Somehow, I managed to finish my Ph.D. (in robotics, not weblogs), and when I moved to UBC in Jan 2004 I swore off blogging. I was done, it was time to grow up and get down to work... and you can see where I've ended up. Why did I come back? I'll save that discussion for tomorrow.

First post...

Robert Sim — Wed, 02 Nov 2005 11:08:00 +0000

Welcome back! It's been two years since I quit blogging cold-turkey. I'll explain my reasons for the hiatus later. In the meantime sit back, relax and enjoy the scenery.

Simra's Mega-Feed

Reading list: Predicting stock prices with HMMs

Tags:

Proof that most methods of making stock market predictions don’t work

MtGox DDOS

Tags:

simra.net returns

How to really make money with Bitcoin

Tags:

How are we doing so far?

Tags:

Stock picks for 3/29

Tags:

The emoticons of pi

Tags:

Vatsal Shah survey of methods for price prediction

Tags:

John Langford on Stock Prediction with ML

Tags:

A first foray into predicting stock prices

Tags:

January in Seattle

January in Seattle

January in Seattle

January in Seattle

January in Seattle

IMG_9009

IMG_9008

IMG_9007

Camping

Hemingway

The bush in our back yard.

Arrested Development pwn Stage 5

Vancouver Folk Music Festival

Main Stage, with a teeny weenie Steven Page rockin' out, Vancouver Folk Music Festival

Veda Hille on Stage 5, Vancouver Folk Music Festival

Bird Man?

Main Stage across the duck pond, Vancouver Folk Music Festival

The unwritten chapter of Alice in Wonderland, Vancouver Folk Music Festival

Wandering Carnival Band, Vancouver Folk Music Festival

Stage 2, Vancouver Folk Music Festival 2009

Another Frightening Show About the Economy

YAMC: Yet another mortgage crisis

Dwight's Perfect Crime

Testing 123

ChChChChanges

Oh crap he's back.

I've moved (again)

Robot hazing

Land disputes in BC

music of the week

Election post-mortem

January in Vancouver

Election coverage...

Happy new year, partie le trois..

Happy new year, part the deux.

Happy new year, part un.

Go green!

Classic breakfasts and NIPS

What's playing...

... the thing about blogs.

Preamble: how on earth did we get here?

First post...