Hello

I have a trial account and downloaded the corn continous future. I found several quite strange quotes (See the screenshots enclosed):
CASE A:
Q1: From 12th July to 15th July the quote drops by 122.50?
Q2: I downloaded corn also from http://www.tradingblox.com . The quote on 12th July is 861.75. Far away from IQ Feed Quote

CASE B:
7 days have quotes of 259, almost 50% below other quotes.

CASE C:
6 days have quotes of 259, almost 50% below other quotes.

CASE D & E
Again, some days with 259 quotes.



How is this possible? I assume that thousdands of users have used this data before and I assume that you do some minimal data checking?

What is then the problem?

Thanks & best regards

seems there is a problem with the PDF, here uploaded again

we are looking into this.

You are looking at the back-adjusted contracts within IQ (where the actual price is adjusted based on the close value of the old and new contract around roll dates). The 259 values are cases where our regular continuation contract (without back adjusting) are incorrect and showing zero values. I have reported this to our market data group for investigation.

I believe this will address all of your issues other than case A which will require some more digging.

We'll keep you updated.

ok, thanks.
BTW:I downloaded the data with NinjaTrader

The issues I pointed out yesterday (all but Case A) have been corrected. There was an issue in our system building the continuous contract data for that contract.

Still looking into Case A

Case A should be resolved by monday.

However, the price values for our Back-Adjusted contracts won't match tradingblox.com because they use a different roll method than we do for continuous contracts.

thanks for the update.

Some questions:
1. About the problem with 259 values. How can I know that you do not have similar kind of problems in many contracts? It is quite time consuming to validate each time series

2. About Case A: You found the bug and resolve on Monday or you still look for the reason? If you found the bug, what was the reason?


I did some more research with unadjusted data, will post it today

I investigated some more:

Compared Unadjusted Corn Quotes with PiFin, Pinnacle and Zenfire
What looks good or is acceptable (for me, who needs EOD data)
- "Open" Quote mostly the same
- Some differences in High/Low/Close quotes. Normally not signifcant, sometimes more than 1%
- Sometimes dates are missing
- Sometimes quotes on a date are constructed (other providers have no quotes, iqfeed has same quote as previous day)

What is not good:
- There are days with 0-values as quotes
- 15th to 31th Dec 2003 no data
- 1st Dec to 12th Dec 2003: Large differences to other sources (between 4.3% and 7.5%)

The problem of price drop on 15th July (CASE A reported at the beginning of this post)
I compared the unadjusted with the adjusted data as well the month contracts (downloaded from turtle page). As you can see in the PDF enclosed there was neither in the July nor in the September contract any movement compared to your backadjusted data series. It looks like that you construct here the backadjusted chain wrongly?

Hello

Is there any update on this topic?

My aplogies for not updating the thread sooner.

The process that calculates all of the back adjustment amounts normally runs on the weekends (this is why I mentioned it should have been fixed on monday). However on monday, when checking it, I noticed there were still some errors in the data. I have been working with our market data group to get the data corrected but it is a fairly time consuming process. So far, the back-adjusted electronic corn contract should be good back through 2000. We are currently working on correcting 1996-1999 and I'll let you know when that process is complete (hopefully this will be later today or tomorrow at the latest).

In response to your specific concerns in the previous post:


This is actually what has been causing problems with the back-adjusted continuous contract. Most of these through 1-1-2000 should be fixed now, please let me know if you find extras.


This could also be releated to the above issue but if you are still seeing these it is also possible that it is by design. IQFeed will show a daily datapoint anytime something trades or settles. If the exchange sent us trades or settlements on the "extra" days then we will likely have a datapoint there. If that doesn't seem to be the case, please let us know some examples and we will look into it.


These should have all been fixed as of my post on Mar 1, 2012 10:25 AM. Let me know if you still see issues.

This should also be corrected at this time.

Can you provide me an example of the differences?

Hope this helps explain things. As I mentioned before, I'll post again as soon as we are finished with the correcting of the back-adjustment back through 1996.