|
First off, to the people disagreeing with me: don't just tell me my conclusions are wrong; tell me where my reasoning is wrong. My attempt here is to explain everything as clearly as possible. All these kids coming in with their own opinions because they've taken some class or read a book or had a little investing experience pretending to know stuff is quite silly. You have zero credibility if you make a claim without an explanation. Also, it's completely pointless if you're off-topic.
Let me better define investing the way I mean it. Investing is growing your money long-term (talking years to decades here) without taking on more risk than you need to. This is not something that you'll actively be spending a lot of time on. It's more of something you set up and just let sit for a long time.
To repeat myself from my last post, you don't know jack squat about what any company is worth. Unless you have a very convincing argument for knowing what you're doing, you're just gambling. You can pick up a book on valuation and learn everything it says, but do you think it's really worth anything when everyone else in the world can also read that book?
What I'm recommending is: Don't bother with valuation and trying to read graphs. Don't bother trying to figure out which companies are good or bad picks. Don't even bother buying individual stocks. I assure you there are already people doing the same thing as you are with way more time, resources, and expertise. It's basically like playing poker against pros and hoping to run good. Seriously, don't kid yourself by thinking you have edge in picking stocks.
Understanding Risk/Return
This part is going to require some basic math/statistics background that I'm going to assume readers are familiar with. People in finance like to talk about investments by describing them with a pair of numbers (µ, σ). Here µ = expected return over one year, and σ = standard deviation over one year. So if a stock is described by (5%, 30%), that means this year you can expect the returns to be between -25% and 35% about two-thirds of the time. It'll be between -55% and 65% about 19 out of 20 years, etc.
If you read my ranting post on EMH, it basically says that you have no idea what µ actually is for any stock. On average, it should be whatever µ is for the stock market. On average, it's equal to whatever rate the economy is growing at, plus you get a little extra for taking on risk. For the past 50 years, µ = 10% a year (including dividends). That looks pretty good. However, factor in inflation (because things cost more every year), it's only 6-7% per year. You also have to consider how much you could have made just by putting it in a savings account. After all that, you're basically looking at 3-4% excess returns. This may not seem like much, but consider that most people work for 40 years or so. $1000 invested at the beginning looks like $3000-4000 by retirement. You're not getting rich quick, but it's a whole lot better than letting it earn interest in a savings account (and way better than keeping it under a mattress).
Diversification
The idea is, that in the long run, the stock market on average reflects the value of the economy. The key ideas here are "in the long run" and "on average". The goal of investing is to realize those long run averages. In the short run though, the stock market can and will fluctuate a lot. You want to minimize the impact of those fluctuations.
Let's take closer look at σ. Suppose somebody offers you a 11:10 coin flip. That is, if it's heads, you win 1.1x what you bet and you lose your bet if it's tails. If you bet $100 on this coin flip, you're either +110 or -100 after one flip. (µ = 5, σ = 105)
Now if two people offered you this bet at the same time, you could bet $50 in each. You'd either end up +110 (1/4 of the time), +5 (1/2 of the time), -100 (1/4) of the time. (µ = 5, σ = 74) How much you expect to win is the same, but your risk is smaller.
If this was offered you three times, it'd look even better. If it was offered to you 10,000 times, and you split your bets evenly, then you wind up with something like (µ = 5, σ = 1).
This is the power of diversification. Instead of coin flips, you can diversify your risk by investing in lots of different stocks. However, the real world is not that rosy. You cannot get rid of all your risks. In general, if the economy is doing well, most stocks will go up. If the economy is doing poorly, most stocks will go down. That is to say, your returns are correlated. At some point you won't be able to lower your σ any more, even if you invested in every single stock in the stock market.
On November 22 2010 14:06 geometryb wrote:
but why isn't a fund not included in the efficient market theory. They can go up or down based on the same principals right?
Of course they are just the same. But because of diversification, regular index funds inherently have lower risks than individual stocks.
Aside 1: to find out σ of the US stock market, there's an index called the VIX. Yes, people actually bet on whether the stock market will get more or less crazy.
Aside 2: Buying stock of the company you work for is pretty silly unless you like risk. Consider the following two options.
Option A - for every day you work at your current job, you pay $1. If you're ever unemployed for 3 months, you receive $1000.
Option B - for every day you work at your current job, you receive an additional $1. If you ever unemployed for 3 months, you have to pay a $1000 fee.
Which one would you take? Most people would probably answer Option A. Yet, Option B is like what happens if you buy stock in the company you work for. If the company does poorly, not only do you lose money on your stock position, but you're also more likely to get laid off. Why would you put all your eggs in one basket?
Timing
Another way to reduce your variance in the stock market is to buy at different times. There's a lot of luck involved with good or bad timing. The problem is you don't really know whether you're getting it in high or low until afterward. It's a good habit to set aside x amount of money to invest every month or every year. In the US, 401(k) plans are pretty good about this and let you auto-invest a portion of your salary every month with no fees.
To sum up, when you're investing, your goal is really just to capture the average return of the stock market with the lowest possible variance. The goal of investing as I'm describing it isn't to "beat the market", but rather "match the market with minimal variance".
As always, feel free to ask any questions.
|
Whao nice! I somehow missed parts 1 and 2... Guess I have some reading for tomorrow 
|
On November 23 2010 00:53 azndsh wrote:
This is the power of diversification. Instead of coin flips, you can diversify your risk by investing in lots of different stocks. However, the real world is not that rosy. You cannot get rid of all your risks. In general, if the economy is doing well, most stocks will go up. If the economy is doing poorly, most stocks will go down. That is to say, your returns are correlated. At some point you won't be able to lower your σ any more, even if you invested in every single stock in the stock market.
You may want to add / make clear, that by diversifying your portfolio you eliminate the unsystematic risk / individual risk of the stocks.
The market risk / systematic risk can not be diversified. I know that you kind of say that in the bold, but I feal like it would be good to differentiate these two.
I don't think I have to explain that any further to you, but I will do so if you like.
And one comment on the returns being correlated. That's why you can lower your risk / eliminate the unsystematic risk by diversifying. With a correlation of 1 (it gotta be 1, thanks to azndsh for pointing that out) that wouldn't be possible.
Again, please ask for further explenation if needed, since I believe you understand what I want to say.
|
Thanks for this! I have never read much or known much about investing and I learned a lot from this =). Lots of stuff I didn't understand before until I read your examples.
|
On November 23 2010 01:19 Copenap wrote:Show nested quote +On November 23 2010 00:53 azndsh wrote:
This is the power of diversification. Instead of coin flips, you can diversify your risk by investing in lots of different stocks. However, the real world is not that rosy. You cannot get rid of all your risks. In general, if the economy is doing well, most stocks will go up. If the economy is doing poorly, most stocks will go down. That is to say, your returns are correlated. At some point you won't be able to lower your σ any more, even if you invested in every single stock in the stock market.
You may want to add / make clear, that by diversifying your portfolio you eliminate the unsystematic risk / individual risk of the stocks. The market risk / systematic risk can not be diversified. I know that you kind of say that in the bold, but I feal like it would be good to differentiate these two. I don't think I have to explain that any further to you, but I will do so if you like. And one comment on the returns being correlated. That's why you can lower your risk / eliminate the unsystematic risk by diversifying. With a correlation of 0 that wouldn't be possible. Again, please ask for further explenation if needed, since I believe you understand what I want to say.
I'm familiar with all the terminology and theory and math, but this is really a high level overview for people who don't have any finance background. We can certainly discuss market vs individual risk, but I don't think it's particularly illuminating for the amount of work it would take to define and explain everything.
The gist of it really is: you can reduce risk by diversifying, but only down to a certain point.
Also you have a typo, it wouldn't be possible with correlation = 1.
|
On November 23 2010 01:34 azndsh wrote:Show nested quote +On November 23 2010 01:19 Copenap wrote:On November 23 2010 00:53 azndsh wrote:
This is the power of diversification. Instead of coin flips, you can diversify your risk by investing in lots of different stocks. However, the real world is not that rosy. You cannot get rid of all your risks. In general, if the economy is doing well, most stocks will go up. If the economy is doing poorly, most stocks will go down. That is to say, your returns are correlated. At some point you won't be able to lower your σ any more, even if you invested in every single stock in the stock market.
You may want to add / make clear, that by diversifying your portfolio you eliminate the unsystematic risk / individual risk of the stocks. The market risk / systematic risk can not be diversified. I know that you kind of say that in the bold, but I feal like it would be good to differentiate these two. I don't think I have to explain that any further to you, but I will do so if you like. And one comment on the returns being correlated. That's why you can lower your risk / eliminate the unsystematic risk by diversifying. With a correlation of 0 that wouldn't be possible. Again, please ask for further explenation if needed, since I believe you understand what I want to say. I'm familiar with all the terminology and theory and math, but this is really a high level overview for people who don't have any finance background. We can certainly discuss market vs individual risk, but I don't think it's particularly illuminating for the amount of work it would take to define and explain everything. The gist of it really is: you can reduce risk by diversifying, but only down to a certain point. Also you have a typo, it wouldn't be possible with correlation = 1.
You're right, of course it's correlation = 1, my bad.
And I agree with your point, that's why I wasn't trying to be a smart ass about it, just wanted to throw that out there and here what you think.
|
What I'm recommending is: Don't bother with valuation and trying to read graphs. Don't bother trying to figure out which companies are good or bad picks. Don't even bother buying individual stocks. I assure you there are already people doing the same thing as you are with way more time, resources, and expertise. It's basically like playing poker against pros and hoping to run good. Seriously, don't kid yourself by thinking you have edge in picking stocks.
Using fundamental analysis or charting is NOT like playing poker against pros for two reasons:
1 - Poker is a zero sum game, whereas the stock market is not. Dividends, stock buybacks, merger & acquisitions, IPOs and positive cashflow growth all point to the fact that the underlying assets of the stock market appreciates even without buyers/sellers.
2 - Even in poker, the pros only play in the highest stakes tables. You can easily find a scrub cash table in Vegas or AC and take $1K on one session on a 1/2 NL table. The same applies to the stock market -- SAC or Greenlight isn't going to be wasting their time in stocks under a few billion in market cap.
And let me assure you, the small and even middle tier hedge fund shops are far from unbeatable, as they're typically running only 5-20 analysts and 1-5 portfolio managers, with which they need to be able to cover thousands of stocks across all sectors.
So yeah, I'm fine with you advocating a diversified buy 'n hold strategy, it's perfectly legit and much easier and takes far less time. But don't talk like fundamental or charting/technical analysis doesn't work. Yes it's harder and more time consuming, but it's definitely do-able -- like playing smaller-stakes poker.
|
On November 23 2010 06:41 Happy.fairytail wrote:
1 - Poker is a zero sum game, whereas the stock market is not. Dividends, stock buybacks, merger & acquisitions, IPOs and positive cashflow growth all point to the fact that the underlying assets of the stock market appreciates even without buyers/sellers.
Markets are zero-sum, its just that the zero point for markets isn't zero, its the growth rate of the economy. I'm not sure this actually impacts your argument much, but its zero-sum in a number of relevant senses.
[edit] And if the harder strategies worked at any tier you wouldn't see pretty much every actively managed fund return less on average than the market does.
|
are there diminishing returns to diversifications? why don't people go out and buy a little bit of every company?
|
They do, via index funds.
|
Fees are charged per transaction, so it's not reasonable to buy 1 share of 1000 firms' stocks instead of 1000 shares of 1 firm's stock. This, incidentally, is why broad ETFs are nice, since they inherently diversify without the messy fees/overhead of mutual funds.
|
On November 23 2010 06:41 Happy.fairytail wrote:Show nested quote +What I'm recommending is: Don't bother with valuation and trying to read graphs. Don't bother trying to figure out which companies are good or bad picks. Don't even bother buying individual stocks. I assure you there are already people doing the same thing as you are with way more time, resources, and expertise. It's basically like playing poker against pros and hoping to run good. Seriously, don't kid yourself by thinking you have edge in picking stocks. Using fundamental analysis or charting is NOT like playing poker against pros for two reasons: 1 - Poker is a zero sum game, whereas the stock market is not. Dividends, stock buybacks, merger & acquisitions, IPOs and positive cashflow growth all point to the fact that the underlying assets of the stock market appreciates even without buyers/sellers. 2 - Even in poker, the pros only play in the highest stakes tables. You can easily find a scrub cash table in Vegas or AC and take $1K on one session on a 1/2 NL table. The same applies to the stock market -- SAC or Greenlight isn't going to be wasting their time in stocks under a few billion in market cap. And let me assure you, the small and even middle tier hedge fund shops are far from unbeatable, as they're typically running only 5-20 analysts and 1-5 portfolio managers, with which they need to be able to cover thousands of stocks across all sectors. So yeah, I'm fine with you advocating a diversified buy 'n hold strategy, it's perfectly legit and much easier and takes far less time. But don't talk like fundamental or charting/technical analysis doesn't work. Yes it's harder and more time consuming, but it's definitely do-able -- like playing smaller-stakes poker.
1 - sure it's non-zero. on average in the long-run it's basically whatever the return rate is for the entire stock market, which is what I'm advocating that people invest in.
2 - what I'm claiming is that it's really hard to beat, and much harder than most people think. furthermore, the amount that you're beating it by almost certainly doesn't justify the amount of additional risk you'd have to take on in order to do so. sure if you enjoy it and you want to do it for a hobby, then that's fine -- just like it's fine to play poker as a hobby.
i'm not writing a guide to investing as a time-consuming and high-risk hobby. i'm writing a guide on how to invest money for retirement.
|
i dont disagree that putting your money into the market for 20-30 years will give you the most consistent return rate. However, this mentality is very dangerous because it is pretty much telling people that investing is easy and requires little to no effort. This will work in a closed environment but when you look at the reality, how many people can leave a good chuck of money and not touch it for 20-30 years? When you buy a house, buy a car, buy anything that is substantial, most people will have to sell their "investments" and by blindly putting your money into an index because it is 'guarantee" can really hurt you.
It may be true that reading annual reports/charts/etc etc may not help you choose a stock that can outperform the market, but it will tell you when the market is crashing/soaring/no movement. Considering how all economist and financial professional look at charts/annual reports, its pretty silly to ignore this stuff completely.
|
On November 23 2010 07:51 kzn wrote:Show nested quote +On November 23 2010 06:41 Happy.fairytail wrote:
1 - Poker is a zero sum game, whereas the stock market is not. Dividends, stock buybacks, merger & acquisitions, IPOs and positive cashflow growth all point to the fact that the underlying assets of the stock market appreciates even without buyers/sellers. Markets are zero-sum, its just that the zero point for markets isn't zero, its the growth rate of the economy. I'm not sure this actually impacts your argument much, but its zero-sum in a number of relevant senses. [edit] And if the harder strategies worked at any tier you wouldn't see pretty much every actively managed fund return less on average than the market does.
Interesting. If the market is a zero-sum game as you say (actually, the word you're looking for is constant-sum game), and if the harder strategies underperform on average, then you're implying the easier strategies outperform on average. But your easier strategies are simply passively buying index funds long-term, so they should only track market performance. So really, if the harder strategies aren't outperforming, and the easier strategies are only performing in line, then who's the outperformer, since the market is constant-sum as you say?
Also, would you mind citing sources or explaining why you think the stock market's constant-sum is just the growth rate of the economy? Macro isn't my strong suit, but I'm pretty sure GDP hasn't been growing 9-10%/annum, whereas the market total return has...
And also, if you have access to long-term hedge fund index returns and cite those, that would be appreciated as well. I can only see the past 3 years on HFR.com.
On November 23 2010 09:12 azndsh wrote:Show nested quote +On November 23 2010 06:41 Happy.fairytail wrote:What I'm recommending is: Don't bother with valuation and trying to read graphs. Don't bother trying to figure out which companies are good or bad picks. Don't even bother buying individual stocks. I assure you there are already people doing the same thing as you are with way more time, resources, and expertise. It's basically like playing poker against pros and hoping to run good. Seriously, don't kid yourself by thinking you have edge in picking stocks. Using fundamental analysis or charting is NOT like playing poker against pros for two reasons: 1 - Poker is a zero sum game, whereas the stock market is not. Dividends, stock buybacks, merger & acquisitions, IPOs and positive cashflow growth all point to the fact that the underlying assets of the stock market appreciates even without buyers/sellers. 2 - Even in poker, the pros only play in the highest stakes tables. You can easily find a scrub cash table in Vegas or AC and take $1K on one session on a 1/2 NL table. The same applies to the stock market -- SAC or Greenlight isn't going to be wasting their time in stocks under a few billion in market cap. And let me assure you, the small and even middle tier hedge fund shops are far from unbeatable, as they're typically running only 5-20 analysts and 1-5 portfolio managers, with which they need to be able to cover thousands of stocks across all sectors. So yeah, I'm fine with you advocating a diversified buy 'n hold strategy, it's perfectly legit and much easier and takes far less time. But don't talk like fundamental or charting/technical analysis doesn't work. Yes it's harder and more time consuming, but it's definitely do-able -- like playing smaller-stakes poker. 1 - sure it's non-zero. on average in the long-run it's basically whatever the return rate is for the entire stock market, which is what I'm advocating that people invest in. 2 - what I'm claiming is that it's really hard to beat, and much harder than most people think. furthermore, the amount that you're beating it by almost certainly doesn't justify the amount of additional risk you'd have to take on in order to do so. sure if you enjoy it and you want to do it for a hobby, then that's fine -- just like it's fine to play poker as a hobby. i'm not writing a guide to investing as a time-consuming and high-risk hobby. i'm writing a guide on how to invest money for retirement.
Cool man, like I already said before, I'm comfortable with your strategy. Let's just leave it at that?
|
On November 23 2010 22:28 Happy.fairytail wrote:
Interesting. If the market is a zero-sum game as you say (actually, the word you're looking for is constant-sum game), and if the harder strategies underperform on average, then you're implying the easier strategies outperform on average. But your easier strategies are simply passively buying index funds long-term, so they should only track market performance. So really, if the harder strategies aren't outperforming, and the easier strategies are only performing in line, then who's the outperformer, since the market is constant-sum as you say?
I don't think there's actually a strategy that outperforms, but the money which seems to be disappearing because there is no such strategy leaks into the salaries/bonuses of the underperforming "hard" strategies.
Hedge funds, for example, leverage themselves like 30 to 1 and then invest in a bunch of stuff. Of course they're going to make a decent return the 9 of 10 years when the market grows - unless you get really unlucky, because they're diversified enough that its vaguely like an index fund most of the time.
But every time they make a profit, they take like 5% of that as a commission. So they're sucking 5% of the profits out of the market, in effect making it perform worse than it appears to. Then, that one year when everything goes to hell, the fund just goes under (or, at best, it doesn't pay any negative bonuses).
In essence, the reason it still works is because people think the harder strategies actually work, put their money in those funds, and flat out leak money to the people managing those funds.
However, I do think there are some private funds that are actually outperforming the market consistently, so I'm not really sure if my argument holds at all times. I'm fairly convinced of it for public funds, however.
Also, would you mind citing sources or explaining why you think the stock market's constant-sum is just the growth rate of the economy? Macro isn't my strong suit, but I'm pretty sure GDP hasn't been growing 9-10%/annum, whereas the market total return has...
That was actually me just being stupid. The constant sum is the growth rate of the market, which takes into account foreign capital flows and whatnot, which boosts it over GDP (i _think_).
And also, if you have access to long-term hedge fund index returns and cite those, that would be appreciated as well. I can only see the past 3 years on HFR.com.
I wish :/ I'm talking from memory of an article I read in the WSJ like 3-4 years ago.
|
i like how you're pulling numbers out of thin air based on some WSJ.com you read and then using flawed reasoning based on that to arrive at absurd conclusions.
|
I like how you haven't made an argument.
|
I would say the most logical follow on from the op is what way do you want to take on extra risk and variance to increase your profit?
Imagine a trade which wins 50% and loses 50%
When you win, you win 160%, when you lose, you lose 40%. +10% ev overall
start with 100 dollars, play 100 times and get an average result, and you get 100*1.6^50*0.6^50 = 12~ dollars - i,e you lost all your money
Where did it go?
Run the simulation for even longer and you will lose even more money
Despite the fact that its a +EV trade
I think this example is the most natural follow up from the OP - following the market and minimising your variance, it is next to important to understand how variance and risk plays a significant role in even your long term expectations.
edit: i liked the second post haha. I feel it is slightly unfair though
Most traders have to take on some kind of risk to make more money.
You cannot always perfectly hedge off every deal you do, so it is still important to fundamentally understand a market
|
So would a "best of breed" strategy work just as well? Best of breed meaning, selecting companies with the best balance sheets and best products poised to be successful in the next coming years. Buying an index funds means that not only am I buying AAPL but also RIMM and a slew of other companies on their way out the door. Can't I just accept that I will take on non-systematic risk and rid myself of "loser" stocks? What if I buy into 15-20 best of breed stocks? Won't that reduce my non-systematic risk significantly?
Do you mind covering "alpha" next? I been reviewing my portfolio and it seems that I out-perform the market everyday. If S&P goes down 1%, I'm only down .5% and vice versa. Is it realistic that this performance will last or is this a hot streak?
|
On November 24 2010 15:02 itzme_petey wrote:
So would a "best of breed" strategy work just as well? Best of breed meaning, selecting companies with the best balance sheets and best products poised to be successful in the next coming years. Buying an index funds means that not only am I buying AAPL but also RIMM and a slew of other companies on their way out the door. Can't I just accept that I will take on non-systematic risk and rid myself of "loser" stocks? What if I buy into 15-20 best of breed stocks? Won't that reduce my non-systematic risk significantly?
Do you mind covering "alpha" next? I been reviewing my portfolio and it seems that I out-perform the market everyday. If S&P goes down 1%, I'm only down .5% and vice versa. Is it realistic that this performance will last or is this a hot streak?
if you beat the market by .5% everyday, wouldn't you beat the market by some 100k% every year? i think azndsh's point is that a loser stock's price already reflects that it is a loser and a winning stock's price already reflects that it is a winner. when you buy aapl, the guy you're trading with is betting that it is overvalued. and when you sell RIMM, the guy is betting that it is undervalued. you don't have any more information available to you to evaluate AAPL or RIMM that says either should go up or down. so unless you have information other people dont have, then it doesn't make too big of a difference.
|
|
|
|
|
|
EPT
