Search results

The short version is that you can't prove what you said.

Yeah, and 19 out of 20 traders "eek" as they go down ugly.

Saying "past price" here means past market price as seen on level I, the same price used to draw a standard chart. There are many other sorts of market data available. If you can't prove that your predictions are based solely on past price, then the exercise is pointless.

You also have to prove that your predictions are based solely on past price.

It is impossible to reliably predict whether the price will be higher or lower in one hour based solely on past price using publicly available methods.

Something is effectively random to the extent that its causes are not known in a way that makes it predictable. The chart of price does not reveal any tendency that extrapolates into the future. Instead, future price is random with respect to past price, as is proved by the very low serial...

The privatization could cause another Wall Street bubble, though that's only a temporary solution.

That's fine, yet we both know that in a "real trading environment," 19 out of 20 traders blow up.

If you're trying to put over that price causes itself, that argument was finished off handily by profturf who pointed out the very low serial correlation of price.

It doesn't accomplish much for you to put so much weight on an argument about the imperfection of efficiency theory. No model has "perfection" on either side, regarding the functioning of the market or one's method of approach. It's a red herring issue that no serious model builder would...

Without a model of cause, or its equivalent, there's no way to predict effect. Whatever other indicators you're using are carrying a heavy load.

Welcome to the threads. A low serial correlation only means future price is poorly predicted by past price, not that future price is random.

No one can force you to learn.

The up or down slope of price on a chart is an effect that does not describe its cause. That is why you can't predict if the cause of an existing "trend" will continue in the future just by looking at effects on the chart. Once a reversal has been established, its chart doesn't reveal how far...

If a "trend" does not extend over a minimum range of price, then it can't be distinguished from background noise.

Quite right. There are much better ways to make money than working for it. I granted this to myself one afternoon after three days with no sleep immersed in the work of Fourier and painting continuous functions without a brush. The market was the obvious choice.

What you're saying is that you can tell when the market will reverse, but you can't give a minimum difference in price between the reversal point and the end of the following trend. In other words, you're not saying anything at all.

This innovation could change trading as we know it.

My claim for the theory of efficiency is that it applies with reasonable accuracy. Perfection is not required in the present discussion.

That's incorrect. Efficiency theory has no requirement of a perfect market as one of its functional parameters. Advocates of the theory make frequent comments on actual market conditions.