| RE: Intraday Daten Historie |       |
Hallo, Klausi!
Würde gerne Deine vorgehensweise, Tickdaten, die im ASCII-Format vorliegen in den GlobleServer einzulesen, kennenlernen. Danach läßt sich betimmt ein Weg finden, wie man dieses automatisieren kann.
Es ist nach meinen Erkenntnissen eben so, daß keine Intradaykurse (Daten des Tages als Zeitreihe mit Uhrzeitstempel) eingelesen werden können, außer eben über das GlobleServer-Format (Dateiendung OMX), Server4.0-Format (Dateiendung OMZ) oder der HistoryBank.COM-Datei, obwohl ich in den Hilfen keinen Hinweis auf dergleichen Einschränkungen finden kann, da dort steht (und der hervorgehobene Absatz sollte eigentlich dei Lösung darstellen):
An ASCII price data file that you would want to import is simply a file that contains line after line of price data for one symbol only. Each line in the file contains the price data for one time-frame, either a tick, day, month, etc.
In addition to lines of data, in order to import the file, you must indicate what order the fields in each line of data are in. You can do this in the following ways:
· Put a header line in the data file
· Use one of the default field orders provided in the data file
· Use a DOP file in the data file
Regardless of the method you use, only the following field names are recognized:
* Date * Time * Open or O
* High or H * Low or L * Close or C
* Volume or V * OPENINT or OI * Other*
The only mandatory fields are Date and Close. If a data file has these two fields only, the program reads the data and also uses the closing price as the open, high, and low prices as well. All other fields are left as 0.
Important Any field names inside the DOP file or header line that are not recognized are ignored and the data stored in the corresponding column of the data file is skipped. The purpose of the Other field is to skip any data other than price data in the file, such as the Symbol. You can use as many Other fields as you want; however, the data in the Other fields are not used.
{** ©1987, 1999 Omega Research, Inc. **}
To be able to plot the price data, the ASCII file must follow certain guidelines:
· Each ASCII data file can contain price information for only one symbol. For example, if you follow IBM and Microsoft, you would have two ASCII price data files: one for IBM and one for Microsoft.
· A carriage return must appear after each line of data and after the last line of data in the file. (The optional DOP file must also contain a carriage return at the end of the line.) To identify the end of a data file, open the file in a text editor (e.g., Notepad) and scroll to the very end of the file. If the cursor drops one line below the last line of data, the file was constructed correctly. If the cursor only goes to the end of the last line, press ENTER to insert a carriage return, and save the file.
· The fields in the data file must be tab delimited, space delimited, or comma delimited. You cannot use any other characters to separate the fields of data. Your Omega Research product interprets a comma, space, or tab as the end of one field and the start of another. No other character field separators (delimiters) are recognized.
Note Omega Research recommends the use of commas to separate the fields of data.
· Dates can be presented in Y2K compliant formats and/or Y2K non-compliant format. You can only use slashes or dashes to separate the day, month, & year. And, if you use slashes or dashes, you don’t have to use the zeros (i.e., 01 can be 1, 02 can be 2, and so on).
Y2K Compliant Formats Examples
MMDDYYYY 01271996 or 01/27/1996 or 1-27-1996
DDMMYYYY 27011996 or 27/01/1996 or 27-1-1996
MMYYYYDD 01199627 or 01/1996/27 or 1-1996-27
DDYYYYMM 27199601 or 27/1996/01 or 27-1996-1
YYYYDDMM 19962701 or 1996/27/01 or 1996-27-1
Y2K Non-compliant Formats Examples
MMDDYY 012796 or 01/27/96 or 1-27-96
DDMMYY 270196 or 27/01/96 or 27-1-96
MMYYDD 019627 or 01/96/27 or 1-96-27
DDYYMM 279601 or 27/96/01 or 27-96-1
YYDDMM 962701 or 96/27/01 or 96-27-1
Y2K non-compliant dates are interpreted based on the following algorithm:
Years 00-20 = 2,000 numbers (e.g., 01 = 2001)
Years 21-99 = 1,900 numbers (e.g., 98 = 199
Example of invalid date:
12796
No zero in month (01). You can correct it by adding “/” or “-“ such as 1/27/96 or 1-27-96. In this example, 96 is interpreted as 1996.
· Each price in the file must be in decimal format. For example, say you have a file that contains price data for corn, which trades in eighths. A price like 193^3/8 must be stored as 193.375.
· For intraday data, the time must be stored in 24-hour format: HHMM (with no colon). For example, 1:15pm would be entered as 1305, and 10:00am as 1000.
· The data file cannot contain characters other than letters or numbers. For example, it cannot contain asterisks, dashes, slashes, and so on. Here is a sample line from an ASCII file: 880125,1405,200.375,201,198.375,200.125,15456
· The symbol name can’t be within the data file (unless you use the field Other in the header line or DOP file). The example below shows a portion of an invalid data file, with an incorrect header line:
"Date","O","H","L","C","V","OI"
DIS,03/15/91,30.812,31.062,30.375,30.812,6520,0
The same data file, with the header line corrected:
"Other","Date","O","H","L","C","V","OI"
DIS,03/15/91,30.812,31.062,30.375,30.812,6520,0
· The field names in the optional header line must be in quotation marks (“ and separated by commas, spaces, or tabs. No other characters can be used to separate field names. Below is an example of a header line that tells the program the order of the fields in the data file. The quotation marks separate each word, and the words are further separated by commas. This identifies that they are field names and where one field name starts and ends.
"Date","O","H","L","C","V","OI"
01/02/91,92.56250,93.40625,92.56250,93.40625,189372,231240
01/03/91,93.34375,94.12500,93.25000,93.78125,283731,240907
· There must be at least three lines of data in the file. If a data file only contains two lines, you will receive an error when loading the data.
· In the case of daily data, the file may not have duplicate or non-sequential dates.
· In the case of intra-day data, the file may not contain non-sequential or duplicate times.
{** ©1987, 1999 Omega Research, Inc. **}
|
21.02.2001, 21:10
