Trading non USD Forex pairs to diversify a portfolio?

 

Hi!

I'm considering to build a portfolio of diversified forex pairs to be traded with an EA, my account currency being USD (IC markets and Eightcap as brokers)

E.g. I'm thinking of EURUSD, GBPUSD as main pairs and AUDCAD, NZDCHF for diversification purposes as an example. Though, wouldn't this diversification be only fictive, as profits made on AUDCAD and NZDCHF would be converted to USD anyway (more specifically conversions to/from CAD and CHF would be needed at entry/exit). Hence wondering if trading these two pairs wouldn't be equivalent to trading AUDUSD and NZDUSD directly, i.e. without possible diversification as all pairs would be USD related.

Any thoughts?

 
Thomas110:

Hi!

I'm considering to build a portfolio of diversified forex pairs to be traded with an EA, my account currency being USD (IC markets and Eightcap as brokers)

E.g. I'm thinking of EURUSD, GBPUSD as main pairs and AUDCAD, NZDCHF for diversification purposes as an example. Though, wouldn't this diversification be only fictive, as profits made on AUDCAD and NZDCHF would be converted to USD anyway (more specifically conversions to/from CAD and CHF would be needed at entry/exit). Hence wondering if trading these two pairs wouldn't be equivalent to trading AUDUSD and NZDUSD directly, i.e. without possible diversification as all pairs would be USD related.

Any thoughts?

In essence, trading AUDCAD and NZDCHF is equivalent to trading AUDUSD and NZDUSD, as you rightly pointed out. The reason is that the underlying risk exposure is still against the US dollar, even though the pairs involve other currencies.

To achieve true diversification, you would need to trade currency pairs that do not involve the US dollar, such as EURJPY, GBPCHF, or AUDNZD. By trading these pairs, your portfolio would have exposure to currencies other than the US dollar, potentially mitigating the overall risk in your portfolio.

NOTE: diversification alone does not guarantee profits or protect against losses.

 
Oleksandr Medviediev #:

In essence, trading AUDCAD and NZDCHF is equivalent to trading AUDUSD and NZDUSD, as you rightly pointed out. The reason is that the underlying risk exposure is still against the US dollar, even though the pairs involve other currencies.

To achieve true diversification, you would need to trade currency pairs that do not involve the US dollar, such as EURJPY, GBPCHF, or AUDNZD. By trading these pairs, your portfolio would have exposure to currencies other than the US dollar, potentially mitigating the overall risk in your portfolio.

NOTE: diversification alone does not guarantee profits or protect against losses.

Thank you for your reply. Though, wouldn't trading the suggested pairs (EURJPY, GBPCHF, AUDNZD) lead to the same issue/lack of diversification, as the account currency is USD?

Hence I'm wondering:

how transparent the underlying conversions (from/to USD) are in MT5's historical trade report

- if true diversification wouldn't be only achievable with several accounts (in USD, EUR, JPY, etc...); there would be a diversification for trading purposes, but not for holding purposes (i.e. if the cash on the non USD accounts is at some point transferred back to an USD account) - so the trading profits should be higher than the possible losses due to longer term currency price fluctuations.

Unless I've missed some aspects in CFD trading.

 

Your observations are 100% correct. The conversions to/from USD may not be fully transparent in the MT5 historical trade reports, as the platform likely shows the profit/loss in the account currency (USD) after the necessary conversions.

However, as you mentioned, even with multiple accounts, there would still be a currency risk when transferring funds between accounts or consolidating profits into a single currency account.

Ideally, the trading profits should outweigh the potential currency fluctuation losses over the long run, but this is not guaranteed neither.
 
Oleksandr Medviediev #:

Your observations are 100% correct. The conversions to/from USD may not be fully transparent in the MT5 historical trade reports, as the platform likely shows the profit/loss in the account currency (USD) after the necessary conversions.

However, as you mentioned, even with multiple accounts, there would still be a currency risk when transferring funds between accounts or consolidating profits into a single currency account.

Ideally, the trading profits should outweigh the potential currency fluctuation losses over the long run, but this is not guaranteed neither.

Thanks for your view

 
Thomas110:

Hi!

I'm considering to build a portfolio of diversified forex pairs to be traded with an EA, my account currency being USD (IC markets and Eightcap as brokers)

E.g. I'm thinking of EURUSD, GBPUSD as main pairs and AUDCAD, NZDCHF for diversification purposes as an example. Though, wouldn't this diversification be only fictive, as profits made on AUDCAD and NZDCHF would be converted to USD anyway (more specifically conversions to/from CAD and CHF would be needed at entry/exit). Hence wondering if trading these two pairs wouldn't be equivalent to trading AUDUSD and NZDUSD directly, i.e. without possible diversification as all pairs would be USD related.

Any thoughts?

Diversification is an important part of both trading and investing.  Understanding and utilizing currency pair correlations is vital for diversification. By mixing pairs that are positively, negatively, and neutrally correlated, traders can manage risk more effectively. At the end everything is collerated with USD directly or indirectly.
 

Choosing pairs to diversify is not as simple as choosing different base currency, because correlations varies depending on several factors:

- If you are trading on low timeframes, correlations can vary greatly based on the time of the day.
- If you are trading higher timeframes (H1, H4, D1), correlations can vary based on political/economical/seasonal influences.

If you write you own EAs you could code it to calculate on real time the correlation between pairs.

I run my EAs on 20+ pairs, but before any trade is opened, my EAs will use Pearson Formula to check the correlation with already running trades. This is the formula:

https://en.wikipedia.org/wiki/Pearson_correlation_coefficient

If you run EAs on higher timeframes you have to option to once a day/week check a correlation table and manually choose wich pairs to keep your EAs running.

Pearson correlation coefficient - Wikipedia
Pearson correlation coefficient - Wikipedia
  • en.wikipedia.org
In statistics, the Pearson correlation coefficient ( PCC ) [a] is a correlation coefficient that measures linear correlation between two sets of data. It is the ratio between the covariance of two variables and the product of their standard deviations; thus, it is essentially a normalized measurement of the covariance, such that the result...
 
Rafael Grecco #:

Choosing pairs to diversify is not as simple as choosing different base currency, because correlations varies depending on several factors:

- If you are trading on low timeframes, correlations can vary greatly based on the time of the day.
- If you are trading higher timeframes (H1, H4, D1), correlations can vary based on political/economical/seasonal influences.

If you write you own EAs you could code it to calculate on real time the correlation between pairs.

I run my EAs on 20+ pairs, but before any trade is opened, my EAs will use Pearson Formula to check the correlation with already running trades. This is the formula:

https://en.wikipedia.org/wiki/Pearson_correlation_coefficient

If you run EAs on higher timeframes you have to option to once a day/week check a correlation table and manually choose wich pairs to keep your EAs running.

Certainly relevant comments and interesting approach in your EA, thanks for describing it.

From the answers already received, I understand trading AUDCAD on a USD account would be equivalent to trading AUDUSD directly, but feel certainly free to comment if you've another view on it

NB: sorry for the delay, have not been notified
 
Thomas110:

Hi!

I'm considering to build a portfolio of diversified forex pairs to be traded with an EA, my account currency being USD (IC markets and Eightcap as brokers)

E.g. I'm thinking of EURUSD, GBPUSD as main pairs and AUDCAD, NZDCHF for diversification purposes as an example. Though, wouldn't this diversification be only fictive, as profits made on AUDCAD and NZDCHF would be converted to USD anyway (more specifically conversions to/from CAD and CHF would be needed at entry/exit). Hence wondering if trading these two pairs wouldn't be equivalent to trading AUDUSD and NZDUSD directly, i.e. without possible diversification as all pairs would be USD related.

Any thoughts?

Everything said above is definitely worth considering, However:

You need to think about the swap fees and spreads when doing this diversification. Many many pairs are simply not good to trade because of their high cost. Check these fees in your symbol list and you will understand what I mean. Spreads can sweep your TP and the currency may not give you enough room to generate acceptable profit.

90% of forex pairs in the symbol list are either USD or GBP or EUR or CHF or JPY etc. related pairs. Among remaining 10%, it is very difficult to find a low to mid cost forex pairs that uncorrelated with these.

On the other hand, if you are using the same EA for all these currency pairs, the diversification may simply not work in practice. It is because, firstly, you need to optimize for that pairs as well - it is  difficult concept and requires a tedious work. Besides, it is very difficult to find an EA to perform well for all pairs. Usually most of them are developed for the mainstream currency pairs used in Forex.

Secondly, when you use the same trading algorithm, the response of EA would be same for all currency pairs when risk appetite change in the market. 

In summary, for diversification to work you should have:

1- Uncorrelated currency pairs with low cost and a well optimised EA for all these pairs,

2- Use different trading algorithms for these pairs.

Otherwise, your diversification likely does not work as you expect -theory is good but there are serious problems in the implementation.