The futures contract is expected to generally trade at a premium to the spot price, which can be attributed to the cost of carrying in the Futures pricing formula. We have previously discussed this in the Futures Module, and any deviation from it presents an arbitrage opportunity.

To get the gist of this situation, think about this scenario: when there is arbitrage opportunity between Spot and Futures markets.

Assuming the spot price is 100 and using the futures pricing formula, one can determine that the fair value of a future is 105. Thus, the ‘no-arbitrage spread’ between the two would be 5.

Assuming the future is mispriced at 98, this creates a spread of 7 (105-98) that can be taken advantage of.

Buying the future at 98 and selling the spot at 100 is all one needs to do. When expiry arrives, futures and the spot will be equalized, securing the spread.

Similarly, if the futures are trading for a higher price compared to its fair value, it is possible to capture the spread by shorting the futures and buying the spot.

We have already covered this, and it is straightforward. Nonetheless, executing arbitrage trades between the spot and futures of USDINR is not feasible due to practical reasons; retail traders do not generally have access to the USDINR spot market.

To trade the currency segment, one should look for the Calendar Spread, which is the difference in two different futures contracts expiring over distinct dates. This method eliminates any need to spot-future spreads.

In a calendar spread, you have to determine if the difference between two futures contracts is unwarranted. Generally, the far-dated futures contract should trade higher than the nearest-dated one; for instance, the August one will almost always exceed July’s. Consequently, some spread is expected. However, when it goes above or below what is typical, potential exists for profitable trades.

As of today the USD INR July Futures is trading at 67.3075, and the August contract is trading at 67.6900.

The spread is calculated as the difference between the two futures contract –

67.6900 – 67.3075

= 0.3825

If you believe the spread of 0.3825 is too high, and it should be 0.2000 instead, there is an arbitrage opportunity available to you, with a potential profit of –

0.3825 – 0.2000

= 0.1825

To take advantage of the spread, you need to buy July Futures and sell Aug Futures at the same time.

Long July Futures at 67.3075

Short August Futures at 67.6900

When you set up a trade where you are buying the current expiry month and selling the further term expiry; it is known as a “Future Bull Spread”. Similarly, setting up a trade where you are selling the current month expiry and buying the further month expiry is called a ‘Futures Bear spread’.

Once you have established the ‘Future Bull Spread’, it is important to keep track of the trade and close the position once its spread falls to 0.2000 or below. You can make money in one of two ways:

1.     When the July (long) leg rises and Aug (short) leg falls
2.     When the long leg raises, and the short leg remains unchanged
3.     When the long leg raises, and short leg rises, albeit at a lower rate.
4.     When the short leg falls faster than the long leg
5. When the long leg remains unchanged and short leg falls

Will the spread converge? When and why might it happen? Is one of the scenarios likely to become reality? To answer these questions you must have a good understanding of the spread, and this can be achieved through backtesting. That is a topic that needs to be explored another day. Right now, I am here to share how simple it is to buy and sell spreads from your trading platform.

What if you had the option to purchase or offer the spread directly? For instance, in our prior example, we determined 0.3825 is an overpriced spread. To capture this spread, you would need to place two orders: buying July Futures and selling August futures.

Executing trades in this manner poses a certain degree of risk, as the prices could potentially change before you complete both purchases, meaning the spread you initially observed may no longer be valid.

It is more efficient to buy the spreads in one go, and not have to handle two individual contracts. If you are a Zerodha user, you can use NEST trader to do this. In the coming days, Pi and Kite will also offer this feature.

Look at the part highlighted in red, as you may have realized; this snapshot is from the market watch. Starting from the left –

2.     We make our selection of spreads, then opt for CDS from the dropdown menu to show currency derivatives as the segment.
3.     FUTCUR suggests that, in terms of CDS spreads, our attention is on Future contracts.
4.     USDINR indicates that we are interested in the USDINR contracts.
5. As you can observe, the full view of the dropdown menu is displayed here. We are interested in the July-August spread, which is what has been chosen. Various other spreads exist.

Once we configure the above-market watch, we submit it and the spread is loaded. This is what it looks like.

I have underlined the last traded cost of the spread. As is evident, this specific spread asset points to the spread between July and August contract.

Note – the spread should be trading at 0.3825 and not really 0.3700 right? Why do you think there is a price difference?

I’ll attempt to explain this from my perspective; any comments would be appreciated! Nevertheless, let’s not stray too far from the main issue of trading spreads.

Have a look at the snapshot below –

The market watch has July, August and a spread contract between the two months loaded.

Rather than focusing on the spread contract, let’s consider a Future Bull Spread (purchasing July and selling Aug). Essentially, you would  –

Sell Aug contract at the Bid Rate – 67.6775

Spread = 67.6775 – 67.3100 = 0.3675

Now, if you were to set up a Future Bear Spread, then you essentially –

Sell July contract at Bid Rate – 67.3075

Spread = 67.6850 – 67.3075 = 0.3775

You have two potential avenues based on your specific goals: a future bull or bear spread.

We must ask ourselves what the spread should be priced at. What could be considered: the Future Bull Spread or the Future Bear spread?

I’d say the spread trades almost near the financial middle of the two, 0.3725. However, the present market has only 0.3700 recorded. It’s possible that this discrepancy is due to one of two things — a newer quote not recorded by the terminal or maybe even a lack of liquidity.

A different explanation here could be that the spread itself is mispriced!

Once you have taken the spread instrument into consideration, all that is left to do is select it from your market watch and use F1 and F2 for buying and selling respectively.

When you open the buy order window, this is what you will see

The window already contains the spread details. You can edit the quantity to fit your lot size needs then submit the order.

As simple as that!

## – USD INR Stats

I found it intriguing to look into the statistical data on USD INR pairs; I grabbed the USD INR spot figures from RBI’s website.

To get some insight into the USD INR exchange rate over time, let us analyze the long term chart from July 2008 to July 2016.

Certainly, in the last 8 years there has been a notable increase in the US Dollar’s value compared to the Indian Rupee. It appears logical that our economic status has remained largely stagnant during this time.

Have a look at the daily return plot of the USD INR

We can observe a few interesting parameters from this –

The average daily return of the USD INR pair is 0.025%, with a maximum and minimum of +4.01% and -2.962%. These figures contrast starkly with Nifty 50’s returns, which stands at +3.81% and -5.92%. This comparison clearly demonstrates that USD/INR is much less volatile when compared to any other indices, including Nifty 50 itself – as evidenced by its volatility numbers.

o   Daily Standard deviation (last 8 years) – 0.567%

o   Daily standard deviation (2015) – 0.311%

o   Annualized standard deviation (2015) – 4.94%

It’s evident that the numbers are much lower than those of the Nifty 50, whose daily volatility and annualized volatility are 0.82% and 15.71% respectively.

In addition, I ran a correlation analysis between Nifty 50 and USD INR. Before I give you the result, I wonder what your opinion on the correlation would be.

For those of you who don’t know what correlation is, here is a quick explanation –

Correlation between two variables gives us a sense of how two variables move concerning each other. Correlation is measured as a number which varies between -1 to +1. For example, if the correlation between the two variables is +0.75, then it tells us two things –

1.     The plus symbol ahead of the figure indicates that the two are positively correlated, that is, they both go in the same direction.
2.     The figure provides us a measure of the power of this phenomenon. Essentially, the more it is close to +1 (or -1), the more likely it is that the two variables will move in harmony.
3. There is no relationship between the two variables when the correlation is 0.

From the correlation of +0.75, it is evident that the two variables move not just in the same direction but also together closely. It should be noted that correlation does not suggest how far either of them will move- all it implies is that movement in one direction can be expected for both of them. For instance, if Stock A increases by 3%, a qualitative corollary can be drawn from a +0.75 correlation between stock A and stock B, implying its movement will also be up, though not necessarily by an equivalent percentage.

If Stock A rises above its usual 0.9% daily return, it is likely that Stock B will follow suit and exceed its 1.2% average day by day gain. These stocks have a correlation coefficient of 0.75, suggesting their behaviour is closely linked.

A correlation of -0.75 points to an inverse relationship between the two variables, with a +2.5% increase in Stock A leading to a unknown decrease in Stock B.

It is worth noting that the correlation of a data series makes sense only when it follows a ‘stationary around the mean’ pattern. Basically, this implies that the values should not drift away too much from the average. The daily returns of USD INR displayed in the graph are a good example of this phenomenon.

This daily average return here is 0.025%. In the case of its daily returns, they exhibit the means-reverting property, this being that they tend to stick to their average value even if the returns initially increase or decrease. Such a data series is considered “stationary around the mean”. Stock/commodity/currency returns are typically stationary, while their prices aren’t as they normally display trending behaviour.

When conducting correlation tests, it is imperative to keep in mind that daily returns are stationary, whereas daily prices have a tendency to trend. Consequently, the correlations should be examined on the former and not the latter.

Calculating the correlation between two variables is quite easy, in fact, has just 2 steps –

1.     Calculate the daily returns
2. Use the ‘=Correl’ function in excel.

Press enter, and you get the correlation between the two variables.

The correlation between stock A and Stock B is reciprocal; it is the same as the correlation between Stock B and Stock A.

I hope you’ve had a decent understanding of correlation, its time I repost the question asked earlier.

If you were to guess the relationship between USDINR and the Nifty 50, would it be positive or negative? Leave aside the actual figure; can you tell if one influences the other in any way?

Let us try and deduce this – If the markets (signifying the whole economy) are performing well, they tend to draw investments from overseas. This means dollars enter the nation and need to be converted to Rupee, thus leading to a sale of dollars for Rupees; thereby making the Rupee stronger. This results in USDINR going down while Nifty 50 increases. The same logic can be applied when looked at from a different perspective, where the market drops and USDINR increases.

The inversely correlative nature of the Nifty 50 and the USDINR is evidenced by their 2015 correlation value of -0.12267.