The other currency pairs

EUR, GBP AND JPY

CURRENCY TRADING.

 The other currency pairs

We focused on the USD INR pair extensively over the last few chapters, and we now look into the other currency pairs that are traded in the Indian markets, namely the EUR INR, GBP INR, and JPY INR. The functioning of the other currency pairs is very similar to the USD INR. Think about it this way – you know how the Nifty 50 contracts work, then you pretty much know or are capable of knowing how Bank Nifty works.

Given this, the agenda for this chapter is to quickly run through the contract specifications of the other three crosses available for us to trade. In the 2nd part of this chapter we’ll dwell on some of the common trading techniques, mainly employing technical analysis, and with this we will conclude our discussion on currencies and start looking into commodities.

So let’s get started.

EUR INR

Globally the EUR USD is one of the most actively traded currencies, however we do not have that contract yet in India, but RBI has given the exchanges a nod to list these crosses as well. So I guess it is a matter of time before we have the EUR USD pair along with GBP USD, JPY USD etc. But for now, we do have EUR INR to trade.

The EUR as we know is the currency of the European Union. Unlike other currencies, the EURO is backed by the economy of many European countries and not just one economy.

The EUR INR contract structure is quite similar to the USD INR contract. Here are the key details that you need to know –

Particular EUR INR Remarks

Lot Size € 1,000 In equity derivatives, lot is number of shares, but here it’s an Euro amount

Underlying The rate of Indian Rupee against 1 EUR

Tick Size 0.25 Paise or in Rupee terms INR 0.0025

Trading Hours Monday to Friday between 9:00 AM to 5:00 PM

Expiry Cycle Upto 12 month contracts Note, equity derivatives have an expiry  upto 3 months

Last trading day Contracts trades till 12:30 PM, 2 days prior to the last working day Equity derivatives continue to trade till 3:30 PM of the expiry day

Final Settlement day Last working day of the month

Margin SPAN + Exposure Usually SAPN is about 1.5%, and exposure is around 1%, hence roughly about 2.5% is the overall margin requirement

Settlement Price RBI Reference rate on the day of Final settlement Closing price of spot

So as you see, the contract specifications are similar to that of the USD INR pair. The only difference is that the lot size in EUR INR is € 1,000 as opposed to $1,000 in USD INR.

Let’s see how this would impact the margins; here is the snapshot of the EUR INR futures –

As you can see, the last traded price of the contract is 74.8950, with this we can estimate the contract value –

Contract Value = Lot size * Contract price

= 1000 * 74.8950

=74,895.0

Assuming the margin is approximately 2.5%, the margin should be in the vicinity if Rs.1,870/-, in fact one can use the margin calculator on Zerodha to get the exact value of the margin required.

GBP INR

The GBP INR contract is probably the 2nd most popular currency contract after the USD INR pair. On the contract specification side of things, everything remains the same except for the lot size and the underlying. The underlying is the exchange rate of 1 GBP in Indian Rupees. The lot size is £1,000, which makes the contract value approximately Rs.89,345/- considering the futures is trading at 89.3450 as of 5thAugust 2016.

As you see below, the margin required for this slightly higher compared to the other two contracts  we’ve already discussed–

By the way, did you know in the international markets that the GBP USD pair is also called the ‘Cable’?. So, when you hear a currency trader say he is short cable, he means he is short GBP USD cross.

JPY INR

The JPY INR contracts are a bit tricky compared to the other currency contracts. The lot size is not the usual 1000 units, but 100000 and the underlying here is the exchange rate for 100 Japanese Yen in Indian Rupees.

The test for seasonality

There is often a lot of debate on the seasonality involved in currencies. By seasonality I mean things like “USD INR always goes down in December” or something like “USD INR always goes up a week before expiry”. In fact many people base their trades based on this expectation without actually validating for seasonality. Given this, we thought we should check for the seasonality in currencies, and needless to say we picked the USD INR spot data to run the required test.

** Warning**

The following discussion can get a bit technical, and this is not meant for regular Varsity readers. If you want a direct answer for whether any sort of seasonality exists in the USD INR pair, then the straight forward answer is – no, there is seasonality of any sort across any time frame. With this conclusion you can jump directly to the next section. However, if you have a statistical approach to things then you may just want to read through. Of course, I’ll try my best to keep it brief.

Also, this section is contributed by our good friend Prakash, any queries regarding this should be directed to prakash.lekkala at gmail dot com.

Seasonality in any time series can be checked by employing a statistical test called “Holt Winters test”. A typical Holt-Winters method has 3 components –

• Level
• Trend
• Seasonality

Level: this indicator measures the average change in USD INR on a YOY basis

Trend: This indicator measures the average change in USD INR on a month on month basis

Seasonality: This indicator measures if there is any seasonal impact on price change. For example – USD INR almost always rises in January, and almost always falls in April etc

There are two possibilities for components (level, trend, and seasonality)

• Additive
• Multiplicative

I guess the details of this are beyond the scope of this discussion.

Holt-Winters test for seasonality:

In Holt-Winters test, we check for seasonality in a time series by building a forecast model (let us call it Model 1) and study its residuals. Model 1 does not have any seasonality component inbuilt. We then build another forecast model with a seasonality component (Model 2) and check for the errors of this model.

We compare the errors of both the models and compare to check if model 2 is gives us a better forecast when compared to Model 1. We do this by employing ‘Chi Square’ test to determine if accuracies are better. If Model 2 is statistically better than Model 1 then we conclude that there is some seasonal pattern in data. However, if the accuracies are same for both models or if Model 1 has better accuracy, there is no seasonality in data.

Seasonality results for USD INR

Check for weekly seasonality:

Model 1 (without seasonality component): The best model is (M, N, N) with coefficients 0.9999

This model indicates that weekly data has only level component and no trend component. The coefficient of “level” is 0.9999 i.e. next week’s price is about 0.9999 times this week’s price.

For readers who are aware of Random Walk Theory will be able to appreciate these parameters. The model is suggesting that on a weekly basis USD INR price movement is a random walk.

Model 2 (with seasonality component):  The best model is (M, N, M) with coefficients 0.7 and 0.0786

This model indicates that weekly data has level and seasonality component. The interpretation is that next week’s price is 0.7 times of this week’s price and the remaining price is contributed by seasonality.

Conclusion: Chi square test concluded that there is 100% chance that model 2 accuracy is same as model 1 accuracy i.e. forcing a seasonality model on USD INR isn’t increasing its accuracy.

This can only happen when there is no seasonality in the data. As the data is prepared for weekly analysis, we can conclude that there is no seasonality on weekly a basis.

Monthly seasonality:

Model 1: The best model is (A, N, N) with coefficients 0.9999

Like in the case of a weekly model, model on monthly data also suggests a random walk.

Model 2: The best model is (A, N, A) with coefficients 0.9999 and 0.0001

This model indicates that next month closing price is almost same as this month’s closing price with a small impact of seasonality.

Conclusion: Chi square test concluded that there is a 20% chance that model 2 accuracy is better than model 1 accuracy. In statistical terms, such improvement in accuracy might happen due to randomness, like the window period you choose, the sample data etc.

Typically in statistics, the norm is to look for at least 95% chance that model 2’s accuracy is better than model 1’s to conclude there is seasonality in data. So in case of USD-INR, we can conclude that there is neither monthly nor weekly seasonality.

The last 8 years USD INR spot data for this is taken from RBI WEBSITE.

So the next time you hear someone make a random statement like “the USD INR pair almost always goes down before Christmas”, then you know he is just trying to sound smart with no real insights. 

Classic TA

Think about conducting a fundamental analysis on a company, for example – Hindustan Unliver Limited. Typically, you would study its business, financial statements, corporate governance, study its peers, and perhaps build a financial model to identify if the stock is worth investing in. Fundamental analysis is kind of a straight forward affair when it comes to equities. However when you look at currency pairs, USD INR for example, there a lot more fundamental dimensions – the macro economics of the USA which is dependent on multiple domestic and international factors and the macro economics of India which is again dependent on multiple domestic and international factors. Once you understand these, you need weigh each one of these against another and build a relative view.

Frankly speaking, this is no easy task and not many are capable of doing this. You need to be an economist with a trader’s mindset to pull off quality fundamental analysis on currency pairs. Perhaps, this is the reason why Technical Analysis (TA) is so much more popular when it comes to trading currencies and commodities. As you probably aware, Technical Analysis assumes that the price that you see on the screen discounts everything including all the complex fundamental views that are panning out at the moment. With this assumption you go ahead and analyze the charts and develop a view point.

TA on currencies and commodities works just like it does on equities. If you are not conversant on how to use Technical Analysis, I’d strongly suggest you read through this module on TA.

I’ll post few snapshots of TA based trade setup

The two encircled candles form a classic candlestick pattern called ‘Piercing pattern’. The piercing pattern suggests the trader to go long on the USD INR pair. As you can see, the trade panned out well without triggering the stop loss.

Here is a bearish Marubozu on GBP INR –

Key takeaways from the chapter

1. The underlying for EUR INR is the spot rate of 1 Euro in Indian Rupees.
2. The lot size for EUR INR is €
3. The underlying for GBP INR is the spot rate of 1 GBP in Indian Rupees. GBP INR is the 2nd most traded contract in the currency segment.
4. The lot size for GBP INR is £
5. Internationally GBP USD is also referred to as the ‘Cable’.
6. JPY INR has the highest margin requirement in the currency segment, perhaps due to the higher volatility.
7. Lot size in JPY INR is 100000.
8. The underlying in JPY INR is the rate of 100 Japanese Yen in Indian Rupees.
9. As opposed to popular belief, there is no seasonality in the USD INR pair – either on the weekly basis or on a monthly basis.
10. TA can be applied to currencies just like the way it can be applied to stocks.