Machine learning in trading: theory, models, practice and algo-trading - page 2825

New comment
 
Maxim Dmitrievsky #:

I wouldn't paraphrase if I'm too lazy to look it up

54 minutes of him "leveraging $300k worth of Scottish kiwi currency". Heard a bell, don't know where it is.

And so it goes with everything else.

Yeah, he heard something and got confused about the $300,000... then at 1:08 he read out (apparently in the chat someone found a link to the article) that there was $40m of his own with leverage of 400, totalling about $15bn in one deal. I.e. 1.5 million lots. I think even now NZD can be dropped a lot with that volume, and even more so in the 90's when there were no hamsters from all over the internet/world to fill the stakes with liquidity.

 
elibrarius #:

Yes, he heard something and got confused with 300k.... then at 1:08 he read out (apparently in the chat someone found a link to the article) that there was $40m of his own with leverage of 400, totalling about $15bn in one deal. I.e. 1.5 million lots. I think even now NZD can be dropped a lot with that volume, and even more so in the 90's when there were no hamsters from all over the internet/world to fill the stakes with liquidity.

Hamsters don't trade forex, they trade at market makers (kind of like big casinos) with their sub-brokers. Real Forex is the exchange of national currencies between banks.
Banks that provide access to forex not through official exchange rates always have kitchen divisions for tradduns who want to change actively. And, as a rule, these are separate legal entities that officially have nothing to do with the bank.

Hence specific market making, which has nothing to do with global markets. The same with futures and other derivatives, i.e. exchange trading.

Yes, there are all sorts of liquidity aggregators between these market makers, but it's still kind of a near market.

I cited his blunder to understand that the man is not orientated in the topic

 
elibrarius #:

Honestly answered - do MMs, exchanges, and whales shave hamsters by 1:17:53

I wonder how much influence such algorithms can have on the appearance/disappearance of arbitrage opportunities. In the sense of both purposeful and random influence.

To be honest, I couldn't watch it till the end - it's a bit hard to understand for some reason).

 
Aleksey Nikolayev #:

I wonder how much influence such algorithms can have on the appearance/disappearance of arbitrage opportunities. In the sense of both purposeful and random influence.

To be honest, I couldn't watch it till the end - it's a bit hard to understand for some reason).

You just don't have an antenna
 
Aleksey Nikolayev #:

I wonder how much influence such algorithms can have on the appearance/disappearance of arbitrage opportunities. In the sense of both purposeful and random influence.

To be honest, I couldn't finish watching it - it's a bit hard to understand for some reason).

I also said a bit about arbitration. They seem to arbitrage themselves, because accounts for market makers have a reduced commission (some start-up exchanges have zero commission). Maybe they also sell software. But in general, they started in 2017. We have been sitting here longer)))) Although they hit the very growth of bitcoin and crypto and became successful on it. Let's see how they can make money on a falling market. We can already see that it is harder, as they have started to look towards funds
 
Maxim Dmitrievsky #:
You just don't have an antenna

They probably all have antennas) Some just hide it in their hair like Curly with FTX)

 
By the way, improvement of trading conditions is dictated solely by business strategies of "market makers" and their sub-brokers trying to find their niche in the competitive environment. This is in the presence of a competitive environment as such, which is there. The last thing they think about is hamsters, but they have to attract them with something. For example, by improving trading conditions while reducing their own profit.

What one well-known broker with good conditions and relatively low profit is doing. But within certain permissible limits, of course, as global markets are still not smelling there.
 
Aleksey Nikolayev #:


Alexei, what type of capital curve is it?
You know, random, stochastic, etc...
What types of forecasts can be applied to it
 
elibrarius #:
And he said a little about arbitrage. It seems that they arbitrage themselves, because accounts for market makers have a reduced commission (some start-up exchanges have zero commission).

If they speculate themselves, it is clear that they reduce arbitrage opportunities by taking liquidity. But it seems to say that they earn on the contrary by increasing liquidity and receiving a fixed subscription fee. Theoretically, if liquidity builds up unevenly, it can create relatively long-lived chains of exchanges for arbitrage. However, they then turn out to be a bit insiders as well)

 
mytarmailS #:
Alexei, what type of capital curve is it?
You know, random, stochastic, etc.?
What types of forecasts can be applied to it

Depends on the instruments and TS, I guess. For a deposit in the Soviet Savings Bank - deterministic, for example).

Traditionally, a model of SB with trend (without reinvestment) and geometric SB (with reinvestment) is used. These models are implicitly assumed when using Sharpe for valuation.

1...281828192820282128222823282428252826282728282829283028312832...3498
New comment
Reason: