Maximize money

[Loren Booda]

Say you wanted to realize the greatest gain in wealth using past economic information, starting at a past investment date of your choosing. What is the most profit (e.g., in terms of GNP percentage) you could actualize before this historical data becomes unreliable (when return becomes relatively random) from the very (nonlinear?) effects following the introduction of your speculation?

 

[russ_watters]

Zero, of course. It says so right on the disclaimer for every investment product: past performance is no guarantee of future gain. Otherwise, everyone would make money on the stock market all the time.

 

[Loren Booda]

So if I were able to decide, using knowledge of the past, to sell retroactively all my stocks in the summer of 1929 and invest in land, I would be no better off, on average, than if I held onto them?

 

[turbo]

So if I were able to decide, using knowledge of the past, to sell retroactively all my stocks in the summer of 1929 and invest in land, I would be no better off, on average, than if I held onto them?

You would be better off, but land is a long-term investment and affords you limited liquidity - you can't move it around like money. Your best bet (given perfect hindsight) would be to invest in companies just before they boom, due to sudden investor interest, a major invention, or revolutionary development, then sell out and go on to invest in your next target company. This can make your investments grow very fast - lots of short-term gains.

 

[russ_watters]

So if I were able to decide, using knowledge of the past, to sell retroactively all my stocks in the summer of 1929 and invest in land, I would be no better off, on average, than if I held onto them?

Ooooooh, I misread your initial post. You want to do an exercise in maximizing profit from a past date to the present using data for the period you are looking to generate profit in - not project it to the future. In that case, I don't understand the statement about unreliable data. A stock price is an exact number - take the historical price of a stock and you can calculate exactly how much money you'd have today given a purchase at a given time in the past.

Since this is just an exercise, why can't you just short-sell a company the day before it goes bankrupt and realize a near instant infinite return (no money invested for an arbitrary amount out)?

I guess I don't see the purpose/point of this question... Are you just asking what type of investments are best, historically?

 

[Loren Booda]

I am trying to understand

1. The effect an individual has on the market by participating or not, and if this choice would usually create a nonlinear disparity. Say I decide to invest in pork bellies (over some other investment) today. Is that choice likely to have a chaotic ripple effect eventually in the economy?

and similarly

2. How quickly any statistic would diverge from hypothetical hindsight of market activity, i. e., how radically and quickly projections become unreliable. Say I have a complete record of all financial transactions in history. If I were able to try maximizing profit (a la "time travel") using the record, how might my financial actions interfere with the eventual reliability of the data?

 

[russ_watters]

It depends a lot on how much you would invest and how diversified it would be. One thing though - once the purchase is made, the influence is over.

In the 20s, it was possible for one investor to influence the market if they were big enough. Today it is only really possible for an individual (with a few exceptions, like Bill Gates) to influence one individual stock. The market is just too big.

 

[Crosson]

I knew that the year 2000 aquisition of AOL by TimeWarner was a disaster, and sure enough it cost TimeWarner about 100 billion dollars. I shhould have sold short

I like the part in Forest Gump where Forest invests in some apple company

 

[EnumaElish]

It depends a lot on how much you would invest and how diversified it would be.

I think one way to look at the OP's question is through the LTCM episode. They did not have future data, but they did have a mathematical model, which predicted the future. It's as if they had future data, only simulated. Acting on the model, they made profit-maximizing investment choices. But actual history turned out much different than what the model predicted.

 

[AsianSensationK]

This is quite an old thread, but really interesting. Although I'm not entirely sure if I am interpreting the question correctly.

I think the data starts to become unreliable as people begin to pick up on the fact that you're beating the market with your data. Expectations start to change. If people wind up following your lead, then you've affected the risk and return trade offs. What was once a risky bet is no longer risky anymore, and people start to expect better returns of your picks. This of course has an impact on the value of a company and ultimately its stock price.

I also don't think it'd be possible for you to go completely unnoticed in the marketplace, because if you had future data that helped you beat the market, you'd probably want to borrow to amplify your profits, right? That's where people would start to ask questions.

 

[Astronuc]

I am trying to understand

1. The effect an individual has on the market by participating or not, and if this choice would usually create a nonlinear disparity. Say I decide to invest in pork bellies (over some other investment) today. Is that choice likely to have a chaotic ripple effect eventually in the economy?

If one has a few thousand dollars or owns a few hundred or thousand shares, one will not have much of an impact on the stock market, and similarly for bonds. However, if one has tens or hundreds of millions or billions of dollars, one will have a significant impact on the markets, especially as AsianSensationK indicated others take notice of one's activities.

and similarly

2. How quickly any statistic would diverge from hypothetical hindsight of market activity, i. e., how radically and quickly projections become unreliable. Say I have a complete record of all financial transactions in history. If I were able to try maximizing profit (a la "time travel") using the record, how might my financial actions interfere with the eventual reliability of the data?

If one had records of rising stocks and falling stocks and one could buy low value stocks and sell high valued stocks one could make a lot of money certainly.

There have been extraordinary events in the past. For example, in the 1980's at least two companies, IIRC Santa Fe RR (or their holding company) and either UTC or GD declared one time huge dividends whereby they dumped huge amounts of cash ~$50/share on shareholders of record in order to avoid takeover. In contrast, other stocks have gone ex-dividend, and certainly if one knew that, one could dump the stock then repurchase it at a much lower price.

Around 1986-87, ARCO stock went from $43/share to ~90$/share and the dividend went from $4/share to $4.50/share, which at a purchase price of $43/share was an attractive return. Even after the stock market crash in Oct 1987, ARCO did well. Also, there were other stockes like Quanex that went from ~$2/share to ~$27/share after the stock market crash. So it would have been a great investment if one had been able to anticipate its ascent.

In the mid 1980's, after Pons and Fleischman announced their cold fusion using palladium, the price of palladium (and platinum) soared, and then dropped.

In the mid 90's, there were settelements on some bonds of various defunct railroads (e.g. Erie Lackawanna which along with others had gone bankrupt with the collapse of the PennCentral RR), and those who were holding the bonds made huge amounts of money. Up to that point, those bonds were worthless, but some were able to purchase those bonds for cents on the dollar. The settlement was about 40% of the bonds' face value. The resolution of those bonds took about 20 years though.

There are numerous historical examples of great increase or decrease in the value of various financial instruments.

As they say, hindsight is 20/20.