Data Integrity Issues

Previous  Top  Next

Close Value is Higher Than the High, or Lower Than the Low

This is probably not an error, but the result of your choice of default data presentations. Upon initial setup, you had the option of presenting the actual prices, regardless of where the settlement lies within the trading range, adjusting the high and low to encompass the settlement price, or adjusting the settlement to be within the daily range.

To change the data presentation for an item in your portfolio, edit the portfolio settings through the portfolio manager. For commodities and stocks not included in your portfolios, use Non-Portfolio Chart options from the Options menu. Either way, click the Rounding/Ranging tab and make your choice, as explained in "Making Close in High/Low Range." An on-screen description displays for each choice.

Both the Open and Close are Outside the Trading Range

This problem is rare, but more likely to occur in overseas data. It has been reported specifically in the CAC-40 Index. The exchange did not publish opening prices in the early days of trading, so to keep charting programs from blowing up with the nonexistent open, the close (settlement) was proliferated into the opening field by CSI. The settlement (fourth field) is not necessarily a traded price and can be outside the high-low range.

We consider this to be the most appropriate approach, as our other options were to show an open that is fictitious, such as the high-low average in the opening field or a zero. If you see this occur, i.e., the open equals the close and both are outside the high-low range, then you have the option of using another program to compute an artificial open (coupling in yesterday's close or calculating the high-low average, for example). Check the on-line factsheet's footnote to find out if this or other anomalies are likely in the markets of interest to you.

Distortions in UA Charts or Computed Data Series

Any time there is a distortion in a data series, a problem exists somewhere along the line. It could stem from a software error, a data point error or a damaged database file. Here are our suggestions for pinpointing the culprit.

·

If you are creating a computed contract, try creating a different computed series from the same data. Try a nearest future series if a back-adjusted won't work, for example. If nothing works correctly for that commodity, but other commodities do work, a History Refresh of the specific database file might do the trick. Get it from the CSI website at http://www.csidata.com.

·

If a problem occurs in creating the same computed contract for all commodities, but those commodities are otherwise normal, it is likely a software error. Go to CSI's web site at http://www.csidata.com, then click "Support," then "Upgrades" to see if there is a new upgrade of UA. Download and install the upgrade (if you don't already have it) and try charting your series again. It may be we have already discovered and corrected the problem.

If the problem continues, please e-mail Techsupport@csidata.com with the specific parameters you used in the failed attempts. We'll try to solve the problem ASAP.

·

If you're using a third-party analysis program to chart the data, try charting it directly through UA. If the identical data series will chart with UA, but not in another format, there is likely a programming error or format incompatibility with that software. Call your software supplier for a remedy.

It is usually pretty easy to spot a data error. It may appear as a spike or single data point, either much higher or much lower than the rest of the data. The balance of the chart is severely compressed. If you suspect a data error, please contact CSI with details on the contract or stock in question, the date, the suspected field or fields and your reason for believing it is in error. If the data point is confirmed as an error, a correction will be transmitted to all UA subscribers in their next download. Ask about our rewards for being the first to report an error.

UA Won't Create Delivery Month Codes Like QuickTrieve

It's true, UA does not create QuickTrieve-style delivery month codes (code 55 for nearest future, for example) because its architecture is far too flexible for the rigid guidelines the codes require. UA uses descriptive filenames and chart descriptions to identify individual data series. Please refer to "File and Chart Naming Conventions" for a better understanding of UA's chart and file names.

Back Adjusted Contracts That Roll N-Days Prior to Expiration Don't Roll Out of Current Contract on Time

UA can execute adjustments based on any roll date perfectly throughout the historical past, but when it comes to the current contract that is about to expire, the program has insufficient information for early rolling. Both UA backadjusters depend upon knowing when a given contract rolls by simply reviewing the historical series of all the delivery months in the database for the commodity in question. Neither backadjuster can absolutely determine when the nearest delivery month will expire until after the fact.

This may seem to be a shortcoming in the software, but it is not. All traders should be aware of the rolling conventions expected and effect the rolling requirements on the appropriate day. CSI provides alert calendar information under the "Alerts" tab for quick review. The last trading day is reliably posted for regular review and examination for most commodities followed by UA.

Half-point values do not display for certain commodities such as CME Lean Hogs and Eurodollar

We emulate the exchanges' practice of not displaying insignificant trailing digits for those commodities whose minimum tick is 2.5 points. For example, a CME Lean Hog price of 67.9 1/4 would be presented as 67.92 (the final 5 has been dropped). UA stores the full price values and allows you to display them without truncation if desired. To do so, click "Display as Decimal" on the "Displaying Data" tab of UA's General Settings. The commodities affected by this truncation are: CME - Live Cattle, Lean Hogs, Pork Bellies, Pork Cutouts, Feeder Cattle, Butter, Libor, Eurodollar, Mexican Peso, S.A. Rand and Russian Ruble; MACE - Live Hogs and Live Cattle; Simex - Eurodollar; and Matif - Rapeseed.