Friday, January 23, 2009
The concept of Gambler's Ruin was mentioned in a very insightful book called Fortune's Formula which I thought deserve more scruntiny, both from gambling and investing point of view.
The story goes as follows. Imagine you have some money and you decide to bet a fixed amount in a game where your probability of winning is 50%. Say you have $100, you want to bet $1 on "Big" and "Small" (and in this case, there is no "House Win" like double sixes. Bcos if there is "House Win", your winning probability would be less than 50% which we do not want in this story yet).
So what is the probability that you will lose all your money after some time?
Well it's 100%!
This is intuitively illogical bcos the odds are 50% right? Why should one lose everything? Well the caveat here is "after some time" which is as good as saying "playing forever". If you play forever, you are bound to lose everything, which can be shown mathematically and that's what we are gonna do in this post. If you decide to stop after winning a certain amt of money, then good for you, it's possible that you achieve your goal and leave the casino with some money.
The mathematical proof of why you can expect to lose everything if you play for a long time (or rather forever) goes as follows:
The probability of either losing your money or doubling your money is 0.5. Consider these mutually exclusive cases:
Case 1: Probability of losing all your money after X1 no. of bets = 0.5
Case 2: After X1 no. of bets, you have doubled your money (0.5), but you continue to play for another X2 bets, so probability of later losing everything again = 0.5*0.5 = 0.25
Case 3: After X1 + X2 no. of bets, you double your money yet again, lucky you! (0.5*0.5), but you continue to play another X3 bets, and the probability of later losing everything = 0.5*0.5*0.5 = 0.125
The no. of cases go on and you add up all the probability that you will go broke = 0.5+0.25+0.125+... = 1
So, as long as you bet the same absolute amt, even in an even odds game, you WILL lose everything in the end.
This link allows you to simulate exactly what will happen and I tried it and recorded down that it takes about 20,000 games to go broke if you have $100 and the odds are 50% (which is quite a lot, but still the math is against you). If the house odds just goes up by 5% (ie your probability of winning is 45%), you lose everything in less than 1,000 games.
So what to do? Well the book says that you should follow this formula called Kelly's Formula to decide how much to bet (which is a % of your money rather than a fixed absolute amount). In this case, the formula actually says don't bet though...
So don't keep betting $1 during the usual CNY Big/Small game. Vary your bet size according to Kelly's Formula!
Thursday, January 15, 2009
Value concepts: I subscribe to most value investing concepts like margin of safety, circle of competence, buy things on sale, value-for-money etc. Most of these are very common-sensical and I don't think there is a need to argue with that. If you have no idea what are these, start reading this blog from the first post back in early 2006.
EMH is bullshit: One thing that I do not agree with other value investors would perhaps be the lack of respect for efficient markets. I think that markets are efficient and it's not easy to beat them. The market is the confluence of everybody's viewpoints and this collective wisdom is actually, usually right, at that point in time. However the market also has an investment horizon that is the average of its participants' horizon, which is usually not very long, abt 1 yr I think. This is probably the only reason why value investors get to have an edge, bcos we look at co.s with solid fundamentals that should outlast market's short-termism over time. Still, I don't think it's easy to get much higher than average return of 8%pa.
Chart reading: I used to think that chart reading is pretty much bullshit bcos if you look at past tosses of coins, can you use that to predict the outcome of your next toss? No. But that is what TA is trying to do. However, prices are not like coins and do have some memory so it might predict future prices. But my guess is its predictive power is probably 2-3 days. So, not that useful. The reason why TA can still work is probably self-fulfilling prophecy at work. Lots of participants playing the game with TA and hence prices do bounce off support levels. I did a simple simulation on the comment section of the last post. Basically, it's possible to make positive return using TA but again, you may not beat market returns. Nonetheless, there are pple who can beat the market using TA and I salute them.
Trading rules: I think this is a useful tool but it works only if you fit it to your temperment, style and investment horizon. E.g. cut loss at -10%. Some can be religious and do it everytime. Some cannot, and see -10% become -50% and curse and swear. According to the value doctrine, you should buy MORE when it drops bcos it just got cheaper right? Well it can go even cheaper, like 2008 and 2009 or even 2010. And your initial analysis must be right. ie things have not changed. If things have changed, the stock is no longer at the original intrinsic value that you calculated, then really must cut. Take profit rule sucks I think. If you sell anything with a 20-30% profit, how are you ever going to make the big buck?
Diversification: Again, this is probably where I differ from the guru (ie Buffett). I think this makes a lot of sense. Diversification is said to be the only free lunch in investing. Of course one major shortcoming is that you must have enough capital. Some textbook says around 30 different investments and they must be relatively uncorrelated lah. This is hard, bcos in today's world, everything just follows everything else. Nevertheless, don't put all your eggs in one basket. Yes we have limited time, money etc. You research on this stock so much and buy 1 lot and see it go up 200%. WTF right? But which is more painful, entire savings become zero or missing out 200%? Beware of the Black Swan!
Well, as most would have realized, I don't subscribe to everything on the value investing doctrine. But I think the core of successful investing has to be Graham/Buffett value philosophy. And now is the time to take action as the fire sale is going on!
Monday, January 05, 2009
As we enter the new year, this is the bonus post that hopefully can make us rich in time to come!
Short Name Industry Subgroup Dividend Yield
SINGAP SHIPPING Transport-Marine 10.53
SIA ENGINEERING Commercial Services 10.05
MOBILEONE LTD Cellular Telecom 9.864
SINGAP AIRPORT T Airport Develop/Maint 9.859
DEL MONTE PAC LT Food-Canned 9.586
SINGAP REINSURAN Reinsurance 9.154
SINGAP PRESS HGS Publishing-Newspapers 8.599
CEREBOS PACIFIC Food-Misc/Diversified 8.503
SINGAPORE POST Transport-Services 8.17
SINGAPORE FOOD Food-Misc/Diversified 7.684
SINGAPORE EXCH Finance-Other Services 7.49
HAW PAR CORP Diversified Operations 6.964
SINGAPURA FINANC Finance-Mtge Loan/Banker 6.667
UNITED O/S INSUR Property/Casualty Ins 6.55
HONG LEONG FINAN Finance-Other Services 6.341
PARKWAY HLDGS Medical-Hospitals 5.804
GREAT EAST HOLD Life/Health Insurance 5.791
SING INV FIN Diversified Finan Serv 5.714
CHUAN HUP HLDGS Transport-Marine 5.263
UNITED OVERSEAS Commer Banks Non-US 5.208
RAFFLES EDUCATIO Schools 5.13
APB BREWERIES Brewery 3.168
This is a list of stocks that have never had a single year of negative free cash flow for the past 8 years and hence I would deem that they should be able to pay dividends for the foreseeable future.
As a surprise, high dividend yield is not selected as a criteria for this list but 15 out of 23 stocks have a dividend yield of more than 6% which I think as high considering that with their strong FCF generative ability, the co.s have less probability of cutting their dividends and what you see is likely to be what you can get as dividends. But don’t be too hopeful, co. mgmt can play the conservative card and decide to keep cash and not pay you.
Incidentally, out of the 700 stocks listed on SGX, only 30 stocks had never had a year of negative FCF and in my opinion, these should be part of the truly investable stocks on SGX.
This drives another point that I have made before. Most stocks on SGX are crap and it is not worth spending time investigating and investing in them.
This does not mean that this list represents ALL the investable stocks on SGX. Or that stocks that don’t make it to this list equals crap. Certainly not! It also does not mean that every stock on this list will turn out to be 5-10 baggers in the future. If this happens, I would require a 25% share of your future profits. But probably, I should have made so much money that I wouldn’t be blogging here. Right?
The list is generated with no regard to the qualitative aspect of the various co.s, like mgmt capabilities, co’s past actions, true source of their cashflow, shareholder friendliness etc. I would advise you to do a lot more work on each of these co.s if you are seriously thinking of putting money into them. Names that I would avoid now would be Parkway: seems like that they might run into cashflow problems with their huge expansions. Others would be Singapura Finance, Raffles Education. Singapura Finance: totally don't know what they do. Raffles Education: wary of what they are doing...
It is unlikely that 80% or even 60% of the stocks on this list will turn out to be spectacular investments. Looking at the 10 year performance to date for this list of stocks, a handful of them have actually declined 50% in the past 10 yrs, making them serious lemons!
Well, that’s investing. Something like finding your life partner. You never get 100% of what you want: e.g. no chores, no quarrels, honeymoon every day, guy’s nights out with no curfew, no anniversary Carat upgrade request but Omega watch present every year, yeah dream on... Hopefully, we get some of what we want, and live with some of what we don't want. And things don't turn out too bad.
My hope would be that this list gives similar results.