0947: “Investing”

This forum is for the individual discussion thread that goes with each new comic.

Moderators: Moderators General, Prelates, Magistrates

Tirian
Posts: 1891
Joined: Fri Feb 15, 2008 6:03 pm UTC

Re: 0947: “Investing”

Postby Tirian » Wed Sep 07, 2011 3:41 am UTC

gmalivuk wrote:True. There are some rather unlikely cases where the utility of a specific sized jackpot is exceptionally high due to specific external factors.

I doubt that's often the case for most people who play the lottery, though, if ever.


Back in the day, I spent a year crunching the numbers on the California lottery (which was particularly nice because the Chron would indirectly report on how many tickets were sold for each drawing). I found that there was a clear window of opportunity where the holdover jackpot was large enough to overcome the normal negative expectation but not large enough that so many people played that there would be an overwhelming danger of splitting your winnings. If you found a way to evade having to pay taxes on the jackpot, it even would have been economically justified (and you used to hear stories about consortiums buying up 14 million tickets thinking that they were tax-exempt), but with the taxes it's a sucker bet even under the best of circumstances.

Faranya wrote:See, I always had a problem with people making statements like that. By that logic, anyone who spends disposable income on something they find enjoyable is "wasting" their money...


Well, yes and no. Going to see a movie or buying an ice cream sundae have extrinsic pleasures and returns that you won't get from the mere fantasy of becoming wealthy. To the degree that you see middle and lower income people fighting against progressive taxation, one might even claim that it is costs even more than a dollar to perceive that you have at least a small chance of getting rich.

But ... gosh, even if it were not irrational to play the lottery, it would be more rational yet to get together with a few friends and recklessly invest in penny stocks. Seems to be that there is the same chance of finding the next Microsoft at the ground level, a much better chance (though still very significant) of not getting totally wiped out, and a much longer window of enjoying the few dollars you spend on it. The difference is that the system is designed to punish lottery players but reward investors.

Xami
Posts: 74
Joined: Tue Sep 29, 2009 8:29 pm UTC

Re: 0947: “Investing”

Postby Xami » Wed Sep 07, 2011 4:12 am UTC

If I don't find it enjoyable / can't do it, nobody should do it!
- every old person

ijuin
Posts: 799
Joined: Fri Jan 09, 2009 6:02 pm UTC

Re: 0947: “Investing”

Postby ijuin » Wed Sep 07, 2011 5:46 am UTC

As for living off of investments, let's assume that the amount of money that you need to live acceptably is roughly equivalent to the median wage for Americans. For the sake of argument, let's use $40,000/year as that amount.

Now, also for the sake of argument, let's suppose that you actually have enough self-restraint to consume $40,000 a year even if you actually possess far more than that (barring severe emergencies such as medical bills). I know that this flies in the face of the way that many people behave, but I am using it to illustrate my point.

So, if you then had $2 million in investments at some point in your life, and were getting a 2% rate of return above the rate of inflation (or $1 million at 4%, or $500,000 at 8% which is kind of unlikely), then you would be able to spend $40,000 a year forever without ever depleting your principal--essentially you would be living off of the investment returns rather than cannibalizing the investment itself. In other words, once you become wealthy above a certain amount, you can pretty much STAY that wealthy as long as you don't go on huge spending sprees.

Tabasco
Posts: 13
Joined: Sat Mar 12, 2011 11:05 am UTC

Re: 0947: “Investing”

Postby Tabasco » Wed Sep 07, 2011 7:12 am UTC

Chrisfs wrote:Some (possibly most) rich people start out rich, have access to better schools, and better jobs through connections, and in some cases, can spend large amounts of money frivolously, as they are simply living off returns on a very very large amount of money.


It's good you used the caveat "possibly". I recently read a fantastic book called "The Millionaire Next Door", with some surprising insights. The author did an extensive statistical survey of millionaires in the US, and found that the vast majority were first-generation wealthy. They didn't inherit, and weren't born into wealth; they worked hard and avoided debt, and that's how they became millionaires. If you'll tolerate a side-tangent, the average millionaire lives in a "middle class" house (not extravagant mcmansions), purchases used cars for cash (not brand-new, and not using debt to finance it), and often owns a small business. Surprisingly, most lawyers & doctors are not millionaires, because they start off with massive debt to re-pay, and often their lifestyle (house, car, etc) requires they spend quite a bit of money trying to appear rich. And what about all the people living in giant houses, with Porsches & Hummers outside? They act rich, and may have high income, but still have very little wealth. Great book.
http://www.thomasjstanley.com/pub-books/1/The_Millionaire_Next_Door.html

I'm surprised at all the disbelief at finding a 10% return in the market. Here's a clue: don't do individual stocks, but find mutual funds. Sure, there are some dogs out there, but there are also quite a few funds that average 10% over a 5-10 year time-frame. Yes, that's not immediate, but if you're investing you should have a distant horizon anyway. If you're looking at a time-frame under 5 years, that's gambling. If you want to invest, you must train yourself to think long-term, i.e. decades. In that case, finding a 10% return over the long term isn't so difficult.

Of course, the smartest way to financial independence is to pay off all debt as fast as possible, because not paying Visa 20% is a lot like keeping 20% of your money, once you're on the other side of that relationship.

Also, one poster very wisely asked what might happen if you're unemployed for 2-4 years ... that's where a good emergency fund comes into play. And that comes from saving, and avoiding debt. It does not come from "fun" lottery tickets.

User avatar
Yakk
Poster with most posts but no title.
Posts: 11045
Joined: Sat Jan 27, 2007 7:27 pm UTC
Location: E pur si muove

Re: 0947: “Investing”

Postby Yakk » Wed Sep 07, 2011 1:09 pm UTC

Tabasco wrote:I'm surprised at all the disbelief at finding a 10% return in the market. Here's a clue: don't do individual stocks, but find mutual funds. Sure, there are some dogs out there, but there are also quite a few funds that average 10% over a 5-10 year time-frame. Yes, that's not immediate, but if you're investing you should have a distant horizon anyway. If you're looking at a time-frame under 5 years, that's gambling. If you want to invest, you must train yourself to think long-term, i.e. decades. In that case, finding a 10% return over the long term isn't so difficult.

Read "The Drunkard's Walk" for why the fact that there are lots of mutual funds averaging 10% over 5-10 year time-frames is neither surprising, nor does it mean that it is reasonable to expect anything near an 8% return on investment.

The short answer is, of course there are such funds. They are great marketing. What you do as a seller of investment funds is start 100 of them with some seed capital from whomever you can convince. After a 2-3 year period (edit: added year period), cut the bottom 20%. Repeat. What you'll end up with is a handful of funds that have "averaged 5-10% over the last 10 years" -- yet past performance does guarantee future returns. The actual performance of these funds will be utterly average at any point -- the mutual funds that are marketed to you at any given time will be ones with great past performance. The ones you own will be ones with average performance.

If you think anything else, then you are being successfully used-mutual-fund dealership'd.

Yes, you can get lucky, and get a fund that used to average 5-10% and still does. Yes, you could end up with some insight that the rest of the world doesn't have, and pick a fund that will generate 5-10%.

But anyone who thinks that getting 8% post-inflation is a reasonable goal for long-term investments doesn't understand how huge 8% post-inflation is, or thinks that we are on the verge of grey-goo industrial revolution.

Note that most millionaires are people who are self-funding a retirement. Save an average of 10k per year for 40 years, own a half-million dollar paid-off home (de-leveraged after a 30 year amortization of massive home price growth), and you are a millionaire. Get lucky (demographically) and do so in a rapidly rising market, or be a genius (or use leverage in a rapidly rising market), and you'll be worth multiple millions. Being able to point out causes for why someone is well off doesn't mean that one should expect that those same set of causes will repeat in the future -- that is a pretty classic fallacy when interacting with a chaotic system. The weatherman is really good at explaining what happened in the weather over the last 3 days, yet abysmal at predicting it over the next 3 days. A hint: the "what happened over the last 3 days" bit is a just-so story.
Last edited by Yakk on Wed Sep 07, 2011 3:42 pm UTC, edited 1 time in total.
One of the painful things about our time is that those who feel certainty are stupid, and those with any imagination and understanding are filled with doubt and indecision - BR

Last edited by JHVH on Fri Oct 23, 4004 BCE 6:17 pm, edited 6 times in total.

gillis
Posts: 1
Joined: Wed Sep 07, 2011 1:23 pm UTC

Re: 0947: “Investing”

Postby gillis » Wed Sep 07, 2011 1:26 pm UTC

I wonder if Randall knew when writing the tooltip, that in fact, Einstein lost his Nobel prize money(or most of it, that is) on the stock market.

turf
Posts: 1
Joined: Wed Sep 07, 2011 2:22 pm UTC

Re: 0947: “Investing”

Postby turf » Wed Sep 07, 2011 2:37 pm UTC

After running through the calculations in a few classes, I no longer thought compound interest was magic, but I did find it interesting. What got me, a couple decades ago, were MLM schemes that touted The Magic of Compound Interest.

A guy I worked with asked to come over after work since he said he had a business proposition for me. Great, I thought, I could use some extra cash doing some programming on the side. When he launched into an MLM speil, I got mad and told him there was no way I was going to participate but I would sit there and listen so he could practice giving the speil, but if he used the phrase "The magic of compound interest" just once, I would kick him out. Sure enough, about halfway through, it came up in his slides and he uttered the phrase - he obviously could not help it. I immediately stood up and showed him to the door.

allauthors
Posts: 5
Joined: Mon Jun 30, 2008 4:06 am UTC

Re: 0947: “Investing”

Postby allauthors » Wed Sep 07, 2011 3:34 pm UTC

2% APR compounded monthly (most common compounding period for investments with that rate of return) for 10 years = 1000*(1+.001666667)^120 = 1221.20
2% uncompounded (to demonstrate the value of compounding even though such an investment does not exist that I know of) = 1200
Net addtl. Gain over 10 years from compounding vs non-compounding: 21.20 (not much) an average of an additional $2/year.

2% APR compounded monthly for 20 years = 1000*(1+.001666667)^240 = 1491.33
2% uncompounded = 1400
Net addtl. Gain: 91.33 (Now it's an average of an additional $4.5/year for ALL 20 years)
2% APR compounded monthly for 30 years = 1000*(1+.001666667)^360 = 1821.21
2% uncompounded = 1600
Net addtl. Gain: 221.21 (7.36 more per year for all 30 years)
2% APR compounded monthly for 50 years (retirement) = 1000*(1+.001666667)^600 = 2716.02
2% uncompounded = 2000
Net addtl. Gain: 716.02 (14 more per year for all 50 years)

And if we take the same numbers but with an 8% gain over 50 years)
1000*(1+.006666667)^600 = 53878 vs 5000 non-compounding vs. 1000 stuffed in a matress... I'll take my chances with compounding interest.

Your.Master
Posts: 52
Joined: Mon Jan 08, 2007 10:03 pm UTC

Re: 0947: “Investing”

Postby Your.Master » Thu Sep 08, 2011 6:05 am UTC

Tirian wrote:If you found a way to evade having to pay taxes on the jackpot, it even would have been economically justified (and you used to hear stories about consortiums buying up 14 million tickets thinking that they were tax-exempt), but with the taxes it's a sucker bet even under the best of circumstances.


I have a method! Play Canadian lotteries. It comes as a lump sum and there's no tax. I'm sure other countries are the same way.

User avatar
Eebster the Great
Posts: 2520
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: 0947: “Investing”

Postby Eebster the Great » Thu Sep 08, 2011 9:33 am UTC

There was a guy who ran a full wheel on the Powerball a while back and managed to win a chunk of change. Due to some mechanical error at one of the stations he actually wasn't able to buy one batch of tickets, but he still got the jackpot, which was the important thing (plenty of small prizes too, I'm sure). As I understand it it was even economical after taxes and taking into account the risk of splitting, and even after having to pay his numerous employees, but such occurrences are exceptionally rare (and of course, by the time they occur, the state has already won plenty on the tickets from the past weeks when nobody won).

Tirian
Posts: 1891
Joined: Fri Feb 15, 2008 6:03 pm UTC

Re: 0947: “Investing”

Postby Tirian » Thu Sep 08, 2011 10:59 pm UTC

Do you have a citation for that? Wikipedia suggests that there are conceivably opportunities of this sort for Powerball if you ignore the risk of shared pots (which would not be prudent according to my research), but they don't mention that any have actually happened. I'd be interested to see what their specific business case was.

User avatar
musashi1600
Posts: 117
Joined: Thu Sep 25, 2008 5:08 am UTC
Location: Hawaii

Re: 0947: “Investing”

Postby musashi1600 » Fri Sep 09, 2011 11:05 am UTC

Just wanted to mention that NPR's Planet Money blog put up this strip in a post.
It's Bicycle Repair Man!

feyayeruka
Posts: 100
Joined: Mon Apr 11, 2011 8:22 am UTC

Re: 0947: “Investing”

Postby feyayeruka » Fri Sep 09, 2011 3:19 pm UTC

I invoke rule 34.
Oh yeah, baby! Work that 2% mmmmm

bigjeff5
Posts: 127
Joined: Tue Nov 10, 2009 3:59 am UTC

Re: 0947: “Investing”

Postby bigjeff5 » Fri Sep 09, 2011 5:42 pm UTC

Yakk wrote:If you think anything else, then you are being successfully used-mutual-fund dealership'd.


This is probably the most ignorant and irrational statement I've read all week.

Stocks are not dice. This is not a random roll of 100 dice, where the successful dice are re-rolled until you have five dice that have rolled 4 sixes in a row. You can gamble on stocks, but stocks themselves are not random in any way, shape, or form. There are companies behind the stocks which have equity, profits, and people attempting to increase the equity and profits.

In light of this, dropping the poorest performing stocks is an excellent way of finding the most consistently well-run and profitable companies. It's like choosing the best players for a baseball team - you drop the ones who suck and keep the ones who consistently perform well. You can reasonably expect those who have performed well in the past to continue performing well, barring some unforeseen tragedy (like a hurricane destroying wheat crops, or something).

Your logic amounts to saying Derek Jeter is suddenly going to play like Joe Suckass next season, based on nothing other than "well, the average baseball player plays like Joe Suckass".

It's exactly as stupid as it sounds.

User avatar
Yakk
Poster with most posts but no title.
Posts: 11045
Joined: Sat Jan 27, 2007 7:27 pm UTC
Location: E pur si muove

Re: 0947: “Investing”

Postby Yakk » Fri Sep 09, 2011 5:55 pm UTC

bigjeff5 wrote:
Yakk wrote:If you think anything else, then you are being successfully used-mutual-fund dealership'd.
This is probably the most ignorant and irrational statement I've read all week.
I'm glad for you. It happens to be reasonably close to true.

Given the weak efficient market hypothesis, unless you are way smarter than everyone else playing the stock market, basically everything you talk about has been already factored into the price of the stock right now.
Stocks are not dice.
The value of a stock, minus their current stock price, is a pretty decent random number generator.

It isn't perfect. But something as simple as "I'll buy the best stocks and pull off 8% year over year investment, why doesn't everyone" is silly.
This is not a random roll of 100 dice, where the successful dice are re-rolled until you have five dice that have rolled 4 sixes in a row.

As for mutual fund picks, what I was talking about is pretty heavily cited. People who pick stocks for mutual funds (that sell to consumers at least) are not measurably better at it than the public market price signal as a whole.
You can gamble on stocks, but stocks themselves are not random in any way, shape, or form.

Oh baby. What do you mean by random? You mean "is determined by information far beyond your ability to determine"?

Because yes, stock prices are random by that measure, just like dice. In order to predict if a given stock will go way up, you'd have to know the consumer and producer behavior of the world for the next few years. You'd have to know if joe worker at the company will die of a heart attack before he manages to implement a (disastrous) | (amazing) plan.
There are companies behind the stocks which have equity, profits, and people attempting to increase the equity and profits.

And there is physics behind a die roll that deterministically determines which side it lands on.

The model for the die is far simpler -- it is far easier to roll a die in such a way that you know what side it will land on, than predict all of the factors that influence stock price in the future.
In light of this, dropping the poorest performing stocks is an excellent way of finding the most consistently well-run and profitable companies.

When you drop the poorest performing stocks, you sell stocks you bought high and are now at a lower price. You then reallocate resources to stocks that used to be cheap, and are now more expensive. Others do the same thing.

The fact that people expect the "expensive" stocks to do better has been factored into their price already by other people. That is why their price is high. So your purchase of it, based off publicly available information, already there. If you expect that their price will go up even more than the market does, then your bet is good.
It's like choosing the best players for a baseball team - you drop the ones who suck and keep the ones who consistently perform well.

Not particularly. On a baseball team, you are limited to N players. In a stock portfolio, you are limited to N$. And the price of the players is determined by other people also trying to build a team. And teams of 500 players are perfectly legal (all on the field at the same time).

So it isn't at all like a baseball team.
You can reasonably expect those who have performed well in the past to continue performing well, barring some unforeseen tragedy (like a hurricane destroying wheat crops, or something).

And you can reasonably expect that their level of performance has been factored into their salary already. You only win if their performance is better than their salary.

Look, it even says it in the small print in every mutual fund I've seen advertised or I've bought: past performance does not determine future results. They know this, and they want your money, so they don't want you to know it.
One of the painful things about our time is that those who feel certainty are stupid, and those with any imagination and understanding are filled with doubt and indecision - BR

Last edited by JHVH on Fri Oct 23, 4004 BCE 6:17 pm, edited 6 times in total.

bigjeff5
Posts: 127
Joined: Tue Nov 10, 2009 3:59 am UTC

Re: 0947: “Investing”

Postby bigjeff5 » Fri Sep 09, 2011 7:16 pm UTC

The reason your statement was ignorant and irrational had nothing to do with whatever information you were basing your claims on. Since you've shared what that information is now, I can demonstrate how ignorant and irrational your statements are. All of this information was turned up by a quick search on Wikipedia, it's not exactly hard to find.

Yakk wrote:
bigjeff5 wrote:
Yakk wrote:If you think anything else, then you are being successfully used-mutual-fund dealership'd.
This is probably the most ignorant and irrational statement I've read all week.
I'm glad for you. It happens to be reasonably close to true.

Given the weak efficient market hypothesis, unless you are way smarter than everyone else playing the stock market, basically everything you talk about has been already factored into the price of the stock right now.


The EMH in all forms is hotly contested by both researchers (aka scientists, if that helps you) and investors. Emperical studies do not support strong variants of EMH, and support for Weak EMH is, well, weak.

Warren Buffet doesn't believe in the Efficient Market Hypothesis, but I suppose he's not as smart as you are, and so is probably wrong and doesn't really know anything about investing*. Eh?

Stocks are not dice.
The value of a stock, minus their current stock price, is a pretty decent random number generator.


Uhh, citation please? I can't come up with any logic that works for that statement, but I'm open to the possibility that I am simply wrong.

It isn't perfect. But something as simple as "I'll buy the best stocks and pull off 8% year over year investment, why doesn't everyone" is silly.

Why is it silly? The market still averages about 10% returns, even with the recession taken into account. I'm not talking about mutual funds here, I'm talking about the market as a whole. A well diversified portfolio (fund or just hand picked stocks, it doesn't matter) aught to average in the neighborhood of 10%. That's how averages work.

This is not a random roll of 100 dice, where the successful dice are re-rolled until you have five dice that have rolled 4 sixes in a row.

As for mutual fund picks, what I was talking about is pretty heavily cited. People who pick stocks for mutual funds (that sell to consumers at least) are not measurably better at it than the public market price signal as a whole.
You can gamble on stocks, but stocks themselves are not random in any way, shape, or form.

Oh baby. What do you mean by random? You mean "is determined by information far beyond your ability to determine"?


If you use fake definitions of random, you can twist it in whichever way you like. That's why words have meaning. It's true that "truly random" is a physical impossibility. That weakens your argument, not mine. But no, what I mean by random is the more traditional measure of random "all possible results have an equal possibility of occurring".

If stocks were random throws of the dice, as you seem to suggest, nobody would make money on the market. Everybody would eventually lose. The huge number of successful investors who are able to be consistently successful is empirical evidence that this is not the case. If stocks were random, winners in the market would be exactly the same as winners in the lottery. They are not, however, because there is a vast amount of information available upon which to base your decisions on. We have TV shows that give out analysis of these stocks, for Christ's sake!

Because yes, stock prices are random by that measure, just like dice.

As I have just stated, and as any halfway decent investor knows, no they are not. It may be possible to abstract the market as a whole and get random, but each individual stock price is based on a wealth of readily available information. It is not random.

In order to predict if a given stock will go way up, you'd have to know the consumer and producer behavior of the world for the next few years. You'd have to know if joe worker at the company will die of a heart attack before he manages to implement a (disastrous) | (amazing) plan.


This happens to be pretty easy to estimate, given the history of the company and the products they already have on the market.

There are companies behind the stocks which have equity, profits, and people attempting to increase the equity and profits.

And there is physics behind a die roll that deterministically determines which side it lands on.


The physics behind the die roll are several orders of magnitude more difficult to track than the freely available information about a stock's company. This makes the die roll several orders of magnitude more difficult to predict than the stock. This is how people make money in the stock market.

The model for the die is far simpler -- it is far easier to roll a die in such a way that you know what side it will land on, than predict all of the factors that influence stock price in the future.

Citation please. I'd really like to see you model that, roll the dice, and give me the correct answer before the dice are shown. The best I've heard of is predicting quadrants (not numbers) in roulette for a specific roulette table and a specific roulette dealer. It's certainly possible, but I've never heard of it being done in a practical way. I know I've never been capable of predicting a die roll, but I've been quite capable of predicting a rise in stock for certain companies consistently. Can I do it for all of them? No, but I don't need to. I can get rich on just a few.

In light of this, dropping the poorest performing stocks is an excellent way of finding the most consistently well-run and profitable companies.

When you drop the poorest performing stocks, you sell stocks you bought high and are now at a lower price. You then reallocate resources to stocks that used to be cheap, and are now more expensive. Others do the same thing.


I see where you are going with that, but it's very short term thinking. When looking at a stock's long term average, the initial buy price is next to meaningless. It's the average performance that matters, and if you bought a stock that has actually gone down in the last 10 years then you should probably never buy that stock again, unless there is a very, very good reason for it.

Accidentally buying high and selling low only happens if you aren't evaluating the stocks for a reasonable amount of time.**

The fact that people expect the "expensive" stocks to do better has been factored into their price already by other people. That is why their price is high. So your purchase of it, based off publicly available information, already there. If you expect that their price will go up even more than the market does, then your bet is good.


I don't know why you are making this argument, it only strengthens my case that the stocks are not random, but relatively predictable. I honestly think you are a little confused about how you think stocks work.

It's like choosing the best players for a baseball team - you drop the ones who suck and keep the ones who consistently perform well.

Not particularly. On a baseball team, you are limited to N players. In a stock portfolio, you are limited to N$. And the price of the players is determined by other people also trying to build a team. And teams of 500 players are perfectly legal (all on the field at the same time).

So it isn't at all like a baseball team.


It's called an analogy. I also was not talking about price, I was talking about performance. A little reading comprehension please.

You can reasonably expect those who have performed well in the past to continue performing well, barring some unforeseen tragedy (like a hurricane destroying wheat crops, or something).

And you can reasonably expect that their level of performance has been factored into their salary already. You only win if their performance is better than their salary.


Again, their salary has nothing to do with it. The initial price of a stock has nothing to do with it. It is their performance over the long term that matters. The analogy is apt.

Look, it even says it in the small print in every mutual fund I've seen advertised or I've bought: past performance does not determine future results. They know this, and they want your money, so they don't want you to know it.


That's called the Cover Your Ass (CYA, for short) legal statement. You may not have noticed in this litigious society of ours, but these statements are everywhere. Companies add them even when they are unenforceable (as in the case of waiving a company's responsibility for your safety). It is there to make it clear that they are not guaranteeing your profits in case there is a market crash or some similar event that causes the fund to lose money.

Past performance is still a very good indicator of future results.

*If you don't know who Warren Buffet is, you shouldn't be talking about investing. But just in case, he is currently the 3rd wealthiest man in the world, and has been in the top spot several times over the last few decades. He tends to trade places with Bill Gates and Carlos Slim on that list every few years. None of this makes Warren Buffet right (just more likely to be right), which is the real reason your investing tips are irrational. Science is a continually moving foundation. It rarely (I'm almost confident enough to say never) gives a concrete answer about anything - it simply improves on its previous answer. For something as poorly understood as economics, absolute statements are almost grounds for dismissal, in my opinion. Almost.

**There is some merit to the idea of buying when the entire market is down, but even this can be completely avoided with careful stock selection in a high market. The point is to buy solid stocks that will give consistent returns. This is the time tested method of getting rich slowly. You can try to maximize your returns by playing the meta-market, but your returns likely won't be as good as simply selecting your stocks with an eye toward the long view.

Edited a couple of times for clarity.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25448
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 0947: “Investing”

Postby gmalivuk » Sat Sep 10, 2011 2:16 am UTC

bigjeff5 wrote:
Stocks are not dice.
The value of a stock, minus their current stock price, is a pretty decent random number generator.
Uhh, citation please? I can't come up with any logic that works for that statement, but I'm open to the possibility that I am simply wrong.
When people who know math talk about randomness, we don't necessarily mean completely uniform randomness. And the plot of Apple price changes in this post actually shows that it's even *harder* to predict than if it were something normal like, say, a normal distribution.

If you use fake definitions of random, you can twist it in whichever way you like. That's why words have meaning. It's true that "truly random" is a physical impossibility. That weakens your argument, not mine. But no, what I mean by random is the more traditional measure of random "all possible results have an equal possibility of occurring".
That is most assuredly not what randomness means. Otherwise, while rolling one die is random, taking the sum of two or more magically becomes nonrandom, because some results are suddenly more likely than others.

each individual stock price is based on a wealth of readily available information.
Right. Just like how I can predict a coin flip (in a vacuum, say), if I have readily available information about the initial velocity and rotation speed and height above the ground.

That doesn't mean coins flipped in vacuums cease to be pretty decent random number generators.

So I'm going to go ahead and say that, until I'm confident that you actually understand what "random" means, you have very little to offer to counter Yakk's argument.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

bigjeff5
Posts: 127
Joined: Tue Nov 10, 2009 3:59 am UTC

Re: 0947: “Investing”

Postby bigjeff5 » Sat Sep 10, 2011 5:29 am UTC

gmalivuk wrote:When people who know math talk about randomness, we don't necessarily mean completely uniform randomness. And the plot of Apple price changes in this post actually shows that it's even *harder* to predict than if it were something normal like, say, a normal distribution.


Yeah, I really gaffed there. I don't know why I disagreed with the stock value minus current stock price being random - that is certainly beyond our ability to predict, and thus random like a coin flip. I think I was just being argumentative, which is bad ju ju.

If you use fake definitions of random, you can twist it in whichever way you like. That's why words have meaning. It's true that "truly random" is a physical impossibility. That weakens your argument, not mine. But no, what I mean by random is the more traditional measure of random "all possible results have an equal possibility of occurring".
That is most assuredly not what randomness means. Otherwise, while rolling one die is random, taking the sum of two or more magically becomes nonrandom, because some results are suddenly more likely than others.


I don't understand, if some outcomes are more likely does that not mean the process is less random? It certainly doesn't become non-random, but it is less random than it was, is it not? Are you not averaging out some of the randomness? For example, if I were betting on dice I'd bet on 7 and win 20% of the time (not exactly good, but better than the average - and some numbers only hit 3% of the time). That sounds a lot less random than the single die roll where any die represents a slightly less than 17% chance. Could you please explain where I am off base on this? For a real-life example, craps wouldn't work at all if the different number winnings weren't weighted based on the probability of the outcome - everyone would just put their money on 5-9 and win 80% of the time.

each individual stock price is based on a wealth of readily available information.
Right. Just like how I can predict a coin flip (in a vacuum, say), if I have readily available information about the initial velocity and rotation speed and height above the ground.

That doesn't mean coins flipped in vacuums cease to be pretty decent random number generators.


I guess here is the crux of my apparent lack of understanding: if the results can be predicted, thereby making the toss unnecessary, how is it still a good random number generator? It is deterministic at that point, so how does it avoid becoming a pseudo-random number generator? Or are you saying it's already a pseudo-random number generator and we're just arguing semantics?

To restate and simplify, my basic point was that stocks over the long term are more predictable than a dice roll, especially when given the significant amount of information that can be applied that exists outside the stock market (i.e. potential upcoming products, who's running the show, looming disasters in the company's base of operations, etc.). They are predictable enough that you can pick good stocks and expect a good return over a long period. This is the foundation of long term investing, and it works. Since 1929 (year end 1928 - pre-crash) to present the average return on the market has been just shy of 9%, and that includes The Great Depression and the most recent crash and recession. The numbers for the last decade are low - 4%, but it's still better than inflation and if you go back 20 years it is 10% even with the poor performance of the 2000's. This is long term investing - long term investing averages out to about 9% yearly gains, even when including the two worst crashes in our country's history. Short term gains may be unpredictable, but long term gains are very predictable.

The mutual fund problem seems moot to me for the simple fact that mutual fund performance itself is tracked, and your personal fund returns are tracked (that is, the money you have made on the funds). A fund that has returned 10% a year in the last 10 years has done so regardless of what the individual stocks averaged, and it will be reflected in your account balance. That figure is not an aggregate average of the performance of the stocks in the fund, but the average yearly returns of the fund itself. If the mutual fund performance was based on the averaged stock performance, then the apparent mutual fund performance would not have dropped in 2007 while actual returns would have plummeted. This did not happen, at least not in my mutual fund account, and I've never heard of it happening in anybody else's either. I'd imagine it would have been all over the news if it had. Anybody who owns mutual funds knows that their returns plummeted in '07, and shot back up in '08 and '09 - right in line with the performance figures of the fund. And I'm talking real cash monies here, not theoretical dollars.

So I may have a poor understanding of random (I think it is more likely that we were simply talking past each other), but Yakk's argument still doesn't hold water without a LOT more evidence to back it up than he gave. The proof is in my mutual fund statements, which did not perform at all the way Yakk's argument says they should have.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25448
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 0947: “Investing”

Postby gmalivuk » Sat Sep 10, 2011 8:05 am UTC

bigjeff5 wrote:I don't understand, if some outcomes are more likely does that not mean the process is less random?
No, that just means it's a *weighted* random number generator. A uniform distribution is just one of many choices.

I guess here is the crux of my apparent lack of understanding: if the results can be predicted, thereby making the toss unnecessary, how is it still a good random number generator?
The result can be predicted given certain facts about the coin as it leaves my hand. It can still be a good RNG because 1) it's hard to actually ascertain those facts in practice and 2) there's probably a fair amount of variation in those things even if it's the same person flipping the coin every time.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

bigjeff5
Posts: 127
Joined: Tue Nov 10, 2009 3:59 am UTC

Re: 0947: “Investing”

Postby bigjeff5 » Sat Sep 10, 2011 4:09 pm UTC

gmalivuk wrote:
bigjeff5 wrote:I don't understand, if some outcomes are more likely does that not mean the process is less random?
No, that just means it's a *weighted* random number generator. A uniform distribution is just one of many choices.


I see, well, I stand corrected then.

The result can be predicted given certain facts about the coin as it leaves my hand. It can still be a good RNG because 1) it's hard to actually ascertain those facts in practice and 2) there's probably a fair amount of variation in those things even if it's the same person flipping the coin every time.


It's not always as hard as you'd think. Some physics post-grads in the late 70's built crude miniaturized computers, hid them in their shoes, and proceeded to fleece casinos at the roulette wheel. Given a known starting point (the number at a certain position on the board when the dealer threw the ball) their computer could calculate which quadrant the ball would land in, conferring an up to 44% advantage over the house. From what I understand all their computer really did was give them the most likely quadrant based on pre-computed values, which makes sense given that it was the 70's, so the mini-computers they were able to build likely had a lot less computing power than a calculator-watch.

In any case, you could do the same thing with the coin flip - assuming you have data on the individual flipping the coin you could calculate probabilities for side showing pre-flip or its opposite.

Still though, in a practical sense it's pretty hard, and using devices like these in casinos is now a felony.

User avatar
Eebster the Great
Posts: 2520
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: 0947: “Investing”

Postby Eebster the Great » Sat Sep 10, 2011 7:14 pm UTC

bigjeff5 wrote:It's not always as hard as you'd think. Some physics post-grads in the late 70's built crude miniaturized computers, hid them in their shoes, and proceeded to fleece casinos at the roulette wheel. Given a known starting point (the number at a certain position on the board when the dealer threw the ball) their computer could calculate which quadrant the ball would land in, conferring an up to 44% advantage over the house. From what I understand all their computer really did was give them the most likely quadrant based on pre-computed values, which makes sense given that it was the 70's, so the mini-computers they were able to build likely had a lot less computing power than a calculator-watch.

In any case, you could do the same thing with the coin flip - assuming you have data on the individual flipping the coin you could calculate probabilities for side showing pre-flip or its opposite.

Still though, in a practical sense it's pretty hard, and using devices like these in casinos is now a felony.

Roulette wheels have also come a long way since the 1970s, and I doubt they are still flawed enough to overcome the house edge. But I guess if you go to a shoddy enough casino this might still work.

You're probably better off counting cards in blackjack.

bigjeff5
Posts: 127
Joined: Tue Nov 10, 2009 3:59 am UTC

Re: 0947: “Investing”

Postby bigjeff5 » Sat Sep 10, 2011 10:25 pm UTC

Eebster the Great wrote:Roulette wheels have also come a long way since the 1970s, and I doubt they are still flawed enough to overcome the house edge. But I guess if you go to a shoddy enough casino this might still work.

You're probably better off counting cards in blackjack.


It should still work just fine, the original cheat was based on physics, not some flaw in the equipment. In fact, according to Wikipedia, the method used by Thorp and Shannon (the guys who created the mini-computer) works best on an unbiased wheel. The easy fix for this is to stop all betting before throwing the ball. Without knowing the starting position there is no way to predict where it will land, and the jig is up. It's also a felony to use a device to cheat, so there is a nice incentive not to do this.

There are certainly other things you could do to foil these predictions - like training your dealer to vary the ball speed more than he already naturally does and varying the speed of the roulette wheel, but if a casino were caught doing that they would be shut down by the gaming commission for cheating in a heartbeat, along with a crapload of lawsuits from everyone who played at any of those tables. Not a good plan.

The latest wheels are much lower profile, causing the ball to drop more slowly, which greatly reduces the influence of any imbalance in the wheel. This helps with wheel bias, for which "cheating" is perfectly legal, but it doesn't do anything to help with the computer cheat. That's one reason there are a dozen cameras on each table, looking for cheaters.

And yeah, you're still probably a lot better off counting cards in blackjack, unless you are lucky and find a really bad roulette wheel - then you can clean up.

E_H
Posts: 27
Joined: Fri May 06, 2011 9:16 pm UTC

Re: 0947: “Investing”

Postby E_H » Sat Sep 10, 2011 10:39 pm UTC

Curiously, 2% is the expected long-term rate of return on stocks after inflation, taxes, and taxes on inflation. Unfortunately, after deducting transaction costs to keep the portfolio representative of the stock market and to reinvest dividends, the rate is even lower, by 0.5% - 2%. If you invest when stock prices are high, though, it may be decades before you break even, and even investing just before a boom will not give high real returns.

For instance, a $1000 investment in the S&P500 in 1950 would have grown to about $6000 by 1980, but the real value in 1950 dollars would be only about $2000. The nominal gain would be $5000, but after tax that would be $3600. You would have $4600 in 1980 after tax, worth about $1533 in 1950 dollars, for a real rate of return of about 1.4%. Reinvesting dividends would produce a higher rate of return, (nominally the portfolio would be worth about $11,000 in 1980, reinvesting dividends) but transaction costs and taxes would eat into that substantially. Realistically a much larger portfolio (several million $) would be needed to emulate the S&P500 and to have the dividends be sufficient to buy full blocks of stock in all 500 companies. Even so, the transaction costs would be high, and real returns would likely be less than 2.5%.

E_H
Posts: 27
Joined: Fri May 06, 2011 9:16 pm UTC

Re: 0947: “Investing”

Postby E_H » Sat Sep 10, 2011 11:12 pm UTC

"Thorp and Shannon (the guys who created the mini-computer) "
Those weren't post-grads, they were professors. Not just professors, but titans. Ed Thorp went on to invent the hedge fund (after also inventing card-counting at blackjack). His fund averaged over 20% return a year after fees while remaining market neutral (no exposure to market movements up or down). See his site for some very informative articles on statistics and investing, (http://www.edwardothorp.com/id10.html) especially his work on the Kelly criterion, which defines the optimal portion of one's bankroll to bet given the amount of edge you have and the odds offered, which generalizes to investments and portfolios of investments.

Claude Shannon invented information theory. In fact, he defined information (-log2[probability]) and introduced the term "bit" as a unit of information.

bigjeff5
Posts: 127
Joined: Tue Nov 10, 2009 3:59 am UTC

Re: 0947: “Investing”

Postby bigjeff5 » Sat Sep 10, 2011 11:40 pm UTC

That was an eye opener!

I used the broader market average as the basis for my calculation and came up with $2,891 (including dividends as well as price changes). I only included the capital gains taxes for 1980, not any other maintenance fees, so this will be lower by a percentage point or two per year, which would probably bring it in line with your figures.

These figures vary a lot, too. Holding out to 1981 your total returns drop a little, but capital gains is 7% lower so you come out ahead in total dollars. However, inflation was a lot higher in '81, so your actual buying power is lower by about 100 dollars. >.<

30 years clearly isn't enough.

Holding until 2010 does a lot better - $1,000 turns into $419,030. Capital gains is only 15% so your total dollar amount is $356,175. That has the same buying power as $38,924 in 1950. Not near as good as it looks, but it is 39 times your initial investment. Inflation kills your gains dead (more correctly inflation has already inflated your gains, and you need to take that into account).

The question is, would you have made more than 39 times that investment if you had spent it elsewhere instead? It seems a lot more possible than it did before.

bigjeff5
Posts: 127
Joined: Tue Nov 10, 2009 3:59 am UTC

Re: 0947: “Investing”

Postby bigjeff5 » Sun Sep 11, 2011 12:02 am UTC

E_H wrote:"Thorp and Shannon (the guys who created the mini-computer) "
Those weren't post-grads, they were professors. Not just professors, but titans. Ed Thorp went on to invent the hedge fund (after also inventing card-counting at blackjack). His fund averaged over 20% return a year after fees while remaining market neutral (no exposure to market movements up or down). See his site for some very informative articles on statistics and investing, (http://www.edwardothorp.com/id10.html) especially his work on the Kelly criterion, which defines the optimal portion of one's bankroll to bet given the amount of edge you have and the odds offered, which generalizes to investments and portfolios of investments.

Claude Shannon invented information theory. In fact, he defined information (-log2[probability]) and introduced the term "bit" as a unit of information.


Yeah I mixed up two different events when I looked up the details of the device. Thorp and Shannon created the first wearable computer in 1961 and it was designed to predict Roulette (apparently the two regularly went gambling together, so a natural progression given their work). They tested it briefly in Vegas, and it worked, but technical difficulties kept them from from using it extensively.

The event I was actually referencing was a group of Santa Cruz post-grad students in the mid/late 70's who created a similar wearable computer and made off with about $10,000 for their effort. There were insulation problems though, which caused shorts that shocked and burned the wearer, which in addition to being painful almost got them caught by the casino. If casinos back then actually were like the movies always portray them, that would not have been a happy ending! So the group called the whole thing off.

Sorry for the confusion.

webgiant
Posts: 250
Joined: Mon Aug 17, 2009 5:36 pm UTC

Re: 0947: “Investing”

Postby webgiant » Sun Sep 11, 2011 8:12 am UTC

mania wrote:
iChef wrote:I've gotten into this argument many times with family members that insist on playing the lottery (The state tax on people who are bad at math).


I disagree with this. It's looking at lottery too linearly. ie you're of the opinion that for every $1 you put in, you get 50c back out.

That's not how it works.

For many, lottery is the only chance they'll ever be truly rich. It gives them a chance at reaching something that they'll never reach through low risk investment and minimum wage alone. Without winning, they'd never be able to afford a yacht for instance.

To ignore both the "fun factor" of gambling and the dream of one day to be living the life of luxury, you've just ignored the entire point of the game. And it is that - a game, a form of entertainment.

Now I'm not saying it's at all a wise investment decision, and I'm not recommending people play. But this "state tax on people who are bad at math", oft-repeated, I feel is incorrect. The players may have actually done the maths, worked out the $10 a week they're spending is not hurting them and the dream of one day having a lot of money (along with the very slim chance of it actually happening) is worth it to them. I mean, on the one hand you have $10 a week * 15 years = $7800 spent on the lottery, with maybe $3900 returned. On the other, you have low risk investment: 7% interest, the same $7800, 15 years - you'd end up around $10k in today's terms. So the dream is maybe costing them $6000 for the 15 years ($10k - $3900). My hobbies - technology and cars - are beyond any doubt going to cost me a lot more than that over the same period. So who am I to judge?


Plus the main point is not that one ticket gives you only a 1 in 17 million chance of winning, its that one ticket raises your chance of winning above zero. Those who do not buy tickets cannot win the lottery (no, I don't care that the spam says someone else bought a lottery ticket for you, they're lying to get into your bank account).

Anonymously Famous
Posts: 240
Joined: Thu Nov 18, 2010 4:01 am UTC

Re: 0947: “Investing”

Postby Anonymously Famous » Sun Sep 11, 2011 1:49 pm UTC

webgiant wrote:Plus the main point is not that one ticket gives you only a 1 in 17 million chance of winning, its that one ticket raises your chance of winning above zero. Those who do not buy tickets cannot win the lottery (no, I don't care that the spam says someone else bought a lottery ticket for you, they're lying to get into your bank account).

Technically, I might. I have a coworker who gives tickets to everyone in the office for Christmas.

User avatar
Yakk
Poster with most posts but no title.
Posts: 11045
Joined: Sat Jan 27, 2007 7:27 pm UTC
Location: E pur si muove

Re: 0947: “Investing”

Postby Yakk » Sun Sep 11, 2011 3:31 pm UTC

Without buying a lottery ticket, you have a non-zero chance of winning.

A winning ticket could blow into your face, and nobody else claims it.

The chance of this happening is low. But, within many standard deviations, it is indistinguishable from buying 10 tickets a week.

To double my chances of winning, I'm in a pool with someone else using this strategy as well.
One of the painful things about our time is that those who feel certainty are stupid, and those with any imagination and understanding are filled with doubt and indecision - BR

Last edited by JHVH on Fri Oct 23, 4004 BCE 6:17 pm, edited 6 times in total.


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: mscha, Rossegacebes and 21 guests