Simulate the situation. If 1,000 people were forced to trade amongst themselves, how would the graph behave? - page 6

 
Maxim Romanov:

You could even do a battle of random robots. Or a whole package of robots with different techniques and decision-making parameters, and see who survives. You could even take some of the robots from the League of Trading Systems. I have a completely random robot. There's random trawl, profit, entry, exit, even the tool is random.

You shouldn't get into such complexities, otherwise there won't be any participants or no one will run them. We should model the market easier, for example, we have the initial price and the volume of assets. Participants know the price and the available volume of assets. To stimulate the participants to make trades, some kind of inflation or tax or penalty is needed - if no trades are made for some time. Thus, the robot is obtained in the form of a function that, if inserted into the class, can easily be used for any number of robots.

The participant either makes a bid: buy or sell at such and such a price, or looks through the bids and accepts some of them.

...And there is no need for any starting price and no volume on the market, let the market itself decide what the price will be. The main thing is that they all have the same amount of money at the beginning.

 
Maxim Romanov:

Why does it need optimization? in the end one of the algorithms will win anyway, it would be interesting to compare which algorithms will win on average with a large number of trials

I don't understand much about such algorithms, so yes, interesting if the algorithm is revealed.

 
Maxim Romanov:

when they do an IPO, how do they put the initial price of the asset, does anyone know?

They bid what they like. What they'd like to sell at.
 
multiplicator:
But speculators on the exchange don't have an asset. And I don't know how many percent of speculators are on the exchange.

I don't know how many percent of the market are speculators. In our case, the market has only speculators who may sell virtual assets to one another and then close the deal. (on each other. or on other bidders).

Like the cryptocurrency market. there, too, the instrument is essentially worthless.

The difference between a crypto and a stock is that if the price of the stock falls, it becomes tempting to buy the plant for next to nothing. But there are risks too, because the company might go bankrupt. With crypto it is the same, it is worthless, but if it does not go bankrupt, it is very profitable to buy $1 bitcoins. So at first sight the assets are different, but not so different. I will make screenshots later with the behaviour of different assets.

 
In general, the best solution in such a situation would be to keep the $100 and quit the experiment.

But some will keep it, and some will decide to trade and make money.

But it would still be better to conduct the experiment so that people would invest their 100 dollars. Then it would be possible to weed out those who do not want to trade, and attract only those who are interested.
 
Maxim Romanov:

Why does it need optimization? in the end one of the algorithms will win anyway, it would be interesting to compare which algorithms will win on average with a large number of trials

I tell them about human psychology and they tell me about robots...

programmers...)
 
multiplicator:
I don't know how many percent of the stock market are speculators.

The market has only such speculators who can sell virtual assets to each other and then close the deal. (on each other. or on other bidders).

Like the cryptocurrency market. there, too, the instrument is essentially worthless.

I don't understand how they can sell without buying or creating - are we modelling a forex scam on the DC principle?

Cryptomanets are an intangible asset that has its own cost (transaction value) and demand (whether you can actually buy something with that currency). In most cases, at the beginning of quoting a coin, the crypto is mined and pamped to sell at a better rate.

 
multiplicator:
I tell them about the psychology of people and they tell me about robots...

programmers...)

With these robots and simulate the psychology: buy cheaper, sell more expensive, fuss when there is no money.

 
multiplicator:
In general, the best solution in such a situation would be to keep 100 dollars and leave the experiment.

But, someone will save, and someone still decides to trade and earn money.

But it would have been better to conduct the experiment so that people would invest their 100 dollars. Then it would be possible to weed out those who do not want to trade, and attract only those who are interested.

Human nature does not allow you to just walk away). Few people probably leave, you should definitely give it a try.

 
I was talking in a thread about three demons - fear, hope and greed.
You could add one more - Euphoria.