Machine learning in trading: theory, models, practice and algo-trading - page 3059
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Where did you get the formula from? From the answer, too?
No, I just put together several currency pairs so that each value has the same weight in pricing.
No, I just put together some currency pairs so that each value has equal weight in pricing.
Interesting.
but it's probably better to base the formula on the stationarity of the series rather than equal weighting, have you tried it?
interesting...
but it is probably better to base the formula on the stationarity of the series rather than on equal weight, have you tried it?
To create a maximally stationary random price chart, you can use so-called "weakly correlated" currency pairs. Such currency pairs have a low level of correlation between each other, which makes them more independent of each other.
Examples of such currency pairs are AUD/USD, USD/CAD, USD/CHF, USD/JPY. These pairs have different features and characteristics, which can contribute to a more stationary index. However, keep in mind that the correlation between currency pairs can change over time and depending on macroeconomic events.
I think that only imagination can limit the outcome here.
To create a maximally stationary random graph
I meant to create a stationary series for arbitrage.
And to create a random series currencies are not needed in principle )
I meant to create a stationary series for arbitrage
AI Answer:
To create a stationary series that can be used for arbitrage, you need to select currency pairs that are highly correlated but may deviate from each other temporarily.
The correlation between currency pairs can be measured using the Pearson correlation coefficient. The closer the coefficient is to 1, the higher the correlation.
Currency pairs with high correlation can include EUR/USD and GBP/USD, USD/CHF and USD/JPY, AUD/USD and NZD/USD, USD/CAD and AUD/USD, USD/JPY and EUR/JPY.
However, for arbitrage, you need to consider not only correlation but also other factors such as the volatility and liquidity of each currency pair, central bank rates and economic news.
I do not use it myself
AI Answer:
To create a stationary series that can be used for arbitrage, you need to choose currency pairs that are highly correlated but can temporarily deviate from each other.
AI is incompetent. Correlation is not important here at all. What is needed is cointegration.
The AI is incompetent. Correlation is not important at all. What is needed is cointegration.
Response from AI (but not ChatGPT):
In order to determine which currency pairs most often correspond to cointegration, you need to conduct a study based on relevant exchange rate data. In general, to determine cointegration between two time series, for example between the exchange rates of two currencies, you need to follow these steps:
Obtain the time series for the two currency pairs you want to analyse.
Perform a unit root test, such as the Dickey-Fuller test, to determine if the series is stationary.
Calculate the cointegration coefficient for the time series, for example, using the Johansen method.
Estimate the proportion of total variation in the time series explained by cointegration, for example, using the coefficient of determination.
If cointegration is found, use the information to make appropriate investment decisions.
The previous reply was from ChatGPT.
AI Answer:
To create a stationary series that can be used for arbitrage, you need to choose currency pairs that are highly correlated but can deviate from each other temporarily.
The correlation between currency pairs can be measured using the Pearson correlation coefficient. The closer the coefficient is to 1, the higher the correlation.
Currency pairs with high correlation may include EUR/USD and GBP/USD, USD/CHF and USD/JPY, AUD/USD and NZD/USD, USD/CAD and AUD/USD, USD/JPY and EUR/JPY.
However, arbitrage requires taking into account not only correlation but also other factors such as the volatility and liquidity of each currency pair, central bank rates and economic news.
I do not use it myself
GBPUSD EURUSD GBPX EURX
Minutes three years old.
Out-of-sample test on new data, minutes three months old
GBPUSD EURUSD GBPX EURX
Minutes are three years old.
Out-of-sample test on new data, Minutes three months old
maybe it would be correct to first write down what was calculated here?
correlation? cointegration? model error?
I have 300 profitable long term trained models, who needs? I can compile the best ones into a bot.
in private, compiled ones are not welcome here. Free of charge, not for commercial use.