Wall Street: Crime of the Century

[Rajiv]

Re: Wall Street: Crime of the Century

I will presume that most of the gold trading today is done electronically -- and few people take physical delivery of the gold! Does that answer your question?

[Jim Nickerson]

Re: Wall Street: Crime of the Century

Rajiv, bart,

The thing you both know but haven't mentioned in trading is the factor of LUCK. Probably the better one is equiped with knowledge the better the odds, but I've yet to believe anyone who says they never lose in trading or investing. I know a guy that is significantly smarter than I am about every aspect of investing who is doing less well than I am this year. For the moment his LUCK is not so hot, and my LUCK is slightly better.

[Rajiv]

Re: Wall Street: Crime of the Century

Jim,

Remember, that my model always worked perfectly in shadow trading -- it was only after a week of real trading that it stopped working when dong actual trades! It still worked if I did no actual trades -- but only shadow trades!

So luck played no role - in my opinion. And you have read a lot of what I have posted -- so you should know my opinions on what truly happened.

[bart]

Re: Wall Street: Crime of the Century

I will presume that most of the gold trading today is done electronically -- and few people take physical delivery of the gold! Does that answer your question?

Thanks, but that's been true for far longer than my prediction charts have existed, and also to the best of my knowledge did not change materially around December 2007.
Perhaps eventually the missing factor or factors will become evident, but so far I've had no luck discovering them and the predictive ability of them is significantly less. :confused:

By the way, the chart is primarily based on various U.S. "hot money" factors plus varying time lags.

[bart]

Re: Wall Street: Crime of the Century

Rajiv, bart,

The thing you both know but haven't mentioned in trading is the factor of LUCK. Probably the better one is equiped with knowledge the better the odds, but I've yet to believe anyone who says they never lose in trading or investing. I know a guy that is significantly smarter than I am about every aspect of investing who is doing less well than I am this year. For the moment his LUCK is not so hot, and my LUCK is slightly better.

No question that luck is a factor, but as Rajiv and I have noted - weird stuff also happens like models breaking for no apparent reason.
Life sure is odd sometimes... and it also beats all the alternatives.

[Rajiv]

Re: Wall Street: Crime of the Century

Look also at the volume of gold being traded -- has that changed drastically? If it has increased over the last 3 years -- then my hypothesis may still be valid -- a version of NSS

Compare the total daily/weekly/monthly volumes to total world gold reserves and see what you come up with!

[Jim Nickerson]

Re: Wall Street: Crime of the Century

Rajiv,

What is NSS?

If I figured out with a model or with just what I can recognize as pertinent conditions suggesting buying or selling something and opened a position using the chosen criteria, and then the price action went against me, then hopefully I would have some mental stop loss point and abandon the trade. If it failed once, I don't think I would abandon the criteria I had used. If it continually failed, I hope I would look for some other criteria.

From what you wrote, it seems you experienced ONE failure and lost real money. If whatever you used continued to work on a trial basis, why not make it fail more than once before abandoning it?

Edit: and I don't think there is any system that cannot fail. It if works more often to your benefit than not, why not use it?

[Rajiv]

Re: Wall Street: Crime of the Century

NSS - Naked Short Selling

Jim,

Model works - when shadow trading - at all times
Model worked for a week - I made money - ergo somebody else lost money (read marketmaker)

Somebody else adjusted trading strategy so that I lost

If I trade model doesn't work - if I do not trade the model works -- there is no subjectiveness in this -- computer follows the market real time

[bart]

Re: Wall Street: Crime of the Century

Look also at the volume of gold being traded -- has that changed drastically? If it has increased over the last 3 years -- then my hypothesis may still be valid -- a version of NSS

Compare the total daily/weekly/monthly volumes to total world gold reserves and see what you come up with!

I have looked at volumes, both of futures themselves (the trend hasn't changed) and at LBMA clearing too - same thing (trend is in place). I've also detrended the data and the correlation is trailing, not leading.

On world gold reserves, I don't have enough confidence in the data (considering things like leasing) to use it. I do use leasing data to give me hints on the equivalent of NSS though.


Here's LBMA gold clearing data, just for grins.

[Jim Nickerson]

Re: Wall Street: Crime of the Century

NSS - Naked Short Selling

Jim,

Model works - when shadow trading - at all times
Model worked for a week - I made money - ergo somebody else lost money (read marketmaker)

Somebody else adjusted trading strategy so that I lost

If I trade model doesn't work - if I do not trade the model works -- there is no subjectiveness in this -- computer follows the market real time

I'm relatively sure there is something about this that I'm not understanding.

Either the model works or it doesn't--regardless of whether or not you have real money at risk.

Use the model and your wife's/girlfriend's money, and keep your money out of the game.

[Rajiv]

Re: Wall Street: Crime of the Century

Either the model works or it doesn't--regardless of whether or not you have real money at risk.

That is what I thought as well -- It appears that that is not how it is!

[Rajiv]

Re: Wall Street: Crime of the Century

Since 2007 you said - it appears that the number of transfers has gone up -- but not the volume of gold in t oz -- but the volume in $ has kept track with the number of transactions -- doesn't that seem to imply a NSS type situation?

[bart]

Re: Wall Street: Crime of the Century

Since 2007 you said - it appears that the number of transfers has gone up -- but not the volume of gold in t oz -- but the volume in $ has kept track with the number of transactions -- doesn't that seem to imply a NSS type situation?

I'm not sure if we're tracking on the actual clearing chart data.

Transfers and ounces traded track quite well together. Its only the dollar volume that has gone roughly flat since early 2006, likely due to the large appreciation in dollar value.

And the trend of both transfers and ounces didn't change in late 2007 when my predictive model broke. Transfers and ounces have continued on their uptrend since then. What am I missing from what you're seeing?

[bart]

Re: Wall Street: Crime of the Century

I'm relatively sure there is something about this that I'm not understanding.

Either the model works or it doesn't--regardless of whether or not you have real money at risk.

Use the model and your wife's/girlfriend's money, and keep your money out of the game.

There's an element that you may not be taking into account - manipulation and similar.

If the model works when money isn't commited and doesn't work when money is committed, then something "unusual" is happening with the market itself.

But I like the concept of not using his own money. ;)

[Rajiv]

Re: Wall Street: Crime of the Century

According to your chart, the yellow has gone flat -- which to me is the ounces of gold traded -- the green appears to be the dollar volume -- the purple the transfers

the purple and green track together while the yellow has gone flat --

What I see is that the amount of physical gold changing hands is flat while the number of transactions and the dollar volume has increased. if you take the ratio of the green line to purple, and the yellow to the purple, what do you see?

Is the volume on the left y axis daily, weekly, monthly, quarterly or annual?