CME Corn and wheat contracts

[aZa]

I am currently looking at the corn and wheat futures prices and creating historical forward curves for both commodities dating back to 1st of January 2002.

Does anyone know answers to the following:

1. Does the exchange add a new contract after each contract's expiration? Or are many new contracts are added at the same time, at a particular time in the year?
2. Precisely which date in the month or year are these new contracts added?

To clarify, consider the information CME's website provides for natural gas: 'A new calendar year will be added following the termination of trading in the December contract of the current year.' no such information is provided for corn or wheat.

Thanks in advance. I hope someone can help.

 

[Mini-Dow]

The CME Group currently allows you to trader corn futures with delivery months up to Dec 2014:
http://www.cmegroup.com/trading/agricultural/grain-and-oilseed/corn_quotes_globex.html

for wheat: http://www.cmegroup.com/trading/agricultural/grain-and-oilseed/wheat.html

Note that the ones close to the 2014 end have low daily volumes, which is a sign of illiquidity. Therefore, most people would trade the Sep/Dec 2011 contracts for now. Hope it helps.

 

[aZa]

Hi Mini-Dow

Thanks for your response. Consider the following page

http://www.cmegroup.com/trading/agricultural/grain-and-oilseed/wheat_product_calendar_futures.html

Under the first trade and last trade column is the information regarding when a new contract was created and until when it will trade. For example, the Sept 11 and Sept 12 contracts came into being on the same day and month a year apart, 15th July 2009 and 15th July 2010 respectively. The last trading is pretty straight forward to work out from the information provided in the 'contract specifications'.

NOW, what I want to know is whether those dates (the day and month) are the same for each contract ( ie March, May, July, Sept, Dec) in every year! As an example, what was the first trading date for the Sept 2002 contract. A reasonable guess would be Friday 14th July 2000 (since the 15th July 2000 is a sat)

When I go back to lets say any contract in the past, I need to know when that contract started trading so that I can get data from that date until the contract's last trading date.

So, basically, how do I work out when a contract started trading. I can't calculate accurate forward curves on any given date if I don't what futures were being traded that date.

 

[amanda.jayne]

Does this help?

http://www.cmegroup.com/trading/agricultural/grain-and-oilseed/corn_contract_specifications.html

Contract Months/Symbols March (H), May (K), July (N), September (U) & December (Z)
Last Trade Date The business day prior to the 15th calendar day of the contract month.

http://www.cmegroup.com/rulebook/cbot-rulebook-listing.html

 

[aZa]

its more when each of the contracts March (H), May (K), July (N), September (U) & December (Z)
start trading. ie when was the first trading date of the July 2004 contract.

Consider you are looking to create the forward curve of corn at date 01/02/2003. Does the July 2004 contract exist? Ignoring the valid liquidity issues raised by Mini-D0w for the moment, the answer is relevant beacuse of the following: on the 01/02/2003 these are the nearby(nrby) contracts
1st nrby March 2003
2dn nrby May 2003
3rd nrby July 2003
4th nrby Sept 2003
5th nrby Dec 2003
6th nrby March 2004
7th nrby ????? July 2004 or Sept 2004 or which contract
8th nrby ??? and so on

 

[Yike Lu]

If you have the data, why not let the data tell you?

 

[aZa]

Hi, thanks for the suggestion. Before posting, I knew that would be my only choice if I failed to get some answers to the questions.
The prob is that the datataset I have may be missing some data points towards the start of the contracts' inception. I can only be sure if I knew the exact start dates.
Anyhow, either I trust the dataset and continue or I try to corroborate it with another source to be sure of its accuracy.
Thanks for your inputs

 

[Yike Lu]

The other thing is that I'm not sure why you need to be very accurate about the very back months? Excluding the effect of the Roll, typically the farther out the contract, the more thinly traded and thus less significant.

Also, typically the day it starts trading is not constant, but the rule stays the same. These markets have been around a long time.

 

[aZa]

I agree completely. In practice I will only be using the first 8 nrbys. I choose these based on the open intereset liquidity. So, in conclusion I was being too pedantic with the dataset analysis.

Anyhow, i'm moving on to create the historical forward this afternoon. later i will be looking at the cointegration of the two curves. The idea is to see what their long term dependence structure is and how far index investment, global politics, food price rise amongst other factors have affected this relationship, both in the short and long term.