From theory to practice - page 1914

 
khorosh:

This is whereTheXpert solves problems over time too).

I'm sure he solved them a long time ago, rather than solving them)

 
Wizard2018:

Lots of formulas and no pictures at all. Nonsense :)

Wrong - there are about twenty charts and a nice university emblem)

 
Aleksey Nikolayev:

Stochastic integrals, in my view, are more needed in portfolio science (risk-neutral portfolios, etc.)

Probably. My experience tells me that for every stochastic integral there is a simple peasant method with 90% similar results :)
 

Asaulenka seems to have moved to a smartphone as well...

I just texted him: "Daddy, do you recognize your son Sasha?" He doesn't say a word, he turns his nose up...

Who treats kids like that, Yura?! (if you're reading these lines)

 
Aleksey Nikolayev:

Eh, Shurik, holy simplicity. Do you really think that no one has thought of "working" with time in financial mathematics before you?

Look for example here, there is a summary of the history of the issue.

Thanks, it's an interesting article.

Imho, the most adequate representation of price dynamics is in tick flow. The closest thing to a tick flow is a RANGE. They are essentially the same ticks but of a larger size. I.e. if we set a task to form from market ticks a 5-digit, for example, 4-digit or 3-digit ticks, then we will obtain the correspondingly sized rendge.

 
secret:
I guess. My experience says that for every stochastic integral there's a simple peasant method with 90% similar results :)

Simplicity is a relative concept) Some people find it easier with Bollinger, others with Ornstein and Uhlenbeck)

As long as it works...

 
Aleksey Nikolayev:

Simplicity is a relative concept) Some people find it easier with Bollinger, others with Ornstein and Uhlenbeck)

As long as it works...

Yes, it doesn't matter what colour the cat is... Some like vodka and some like cognac).

 
sibirqk:

Thanksb, that's an interesting article.

Imho, the most adequate representation of price dynamics is in tick volume. The closest thing to a tick flow display is the Renches. Essentially it is the same ticks but of a larger size. I.e. if the task is to form from market ticks a 5-digit, for example, 4-digit or 3-digit ticks, then correspondingly sized rendges will be obtained.

The main problem is that a tick tells us nothing about the current market density, we try to judge volume indirectly through the number of ticks, thinking that it correlates with the number of ticks. Of course it is, but only partly. For example, 10 points in the DOM may contain 1000 lots of limits or 100, the movement in ticks will be the same, but in the first case 10 points have bought enough, while in the second case it is not enough and requires more, but we cannot see it by ticks and cannot distinguish two principally different situations.

If the amplitude of movement is less than the size specified for the range bars, new bars will not appear, though in reality the chart was moving right on ticks at this point and gave a signal. This is another problem, which partially follows from the first one. Another is a reliable source of tick volumes. We take five different brokers and see that their volumes are very different, there is no deep history and even within one broker the volumes may vary significantly, probably, due to rotation of liquidity providers. Ideally, we need an equivolume chart based not on tick volume but on lots. Forex lots, yeah.

 
sibirqk:

Thanks, interesting article.

Imho, the most adequate representation of price dynamics is in the tick stream. The closest thing to a tick flow is the Renches. Essentially it is the same ticks but of a larger size. I.e. if the task is to form a 5-digit market ticks, for example 4-digit or 3-digit ticks, then the corresponding size will be obtained.

It's great for one asset - there's a huge variety of grandfatherly ways (and the grandchildren are trying too). It would be interesting to summarise all this for portfolios of assets.

 

Off-topic at all. You can ignore it. There is something of a tug-of-war procedure in the price movement, the movement of the rink on the rope relative to the ruler on the ground. The tug-of-war is non-stop 24/5 with hot meals and beds for sleeping. The number of people on both sides is unpredictably variable over time.

There is endless bidirectional movement of people from team to team. All legal movements of an individual participant are described and timed for history and the issuing of rewards. The reward is paid for the contribution made to the team's local win over a period of time. If a contribution has been made and the team loses in the period, it has gone to waste, so has the reward.


Such a pseudo-model.

Reason: