[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: US stock data file conversion



PureBytes Links

Trading Reference Links

Jim, here is another question along the same lines: how to convert ASCII
file data into MSPRO data (appending the data to end of MSPRO files.)
Could not do it with Downloader's Conversion tool. What is a PERL
script?
    I have attached a small part of a Quotes.com (ASCII text) downloaded
file for 3/9/97 NYSE stocks. Attached file is "stks0309.txt"   I tried
to used MSPRO 6.0 Downloader's Conversion Tool.
  Here is the Convert Securities Dialog (p. 65 of manual).
     Source: ASCII Text
     Folder:   e:\downloads
     File Name:  stks0309.txt

    Destination:    File Type:  MetaStock
                                Folder: c:\mspro\apractic
                                FileName:  (grayed out)

     Conversion Report was revealing.  Added: 0    Discarded: 0
    Showed correct source but incorrect destination of
c:\mspro\apractic\stks0309.txt
           The destination should have been c:\mspro\apractic
     Error report:   "Warning #2052 (line 2949). No data from source
file converted. "

                               CMA



> On Fri, 19 Sep 1997, Marco Guglielmo wrote:
> > I am using the daily file quotations downloaded free from the
> > www.gofutures.com site.
> > Considering that the ASCI file has more than 16,000 strings of data,
>
> > does anybody advise me how to extract in a quick way only some
> hundreds
> > stock quotations?
>
> Describe the data more accurately. What do you mean by 16K strings of
> data? Post a sample of the data and also a sample of what you would
> like
> to extract and I can probably post a sample of Perl code which will do
>
> what you want. Alternatively to Perl, you can import all into Access
> and
> then use queries to extract. Again, a sample of data is better than
> any
> description.
>
> Cheers,
>
> Jim


<x-html>AAP,950309,35.0000,35.0000,34.2500,34.3750,23800,0
AAT,950309,12.5000,12.5000,12.2500,12.3750,16700,0
AB,950309,37.7500,37.7500,37.5000,37.5000,18700,0
ABE,950309,7.6250,7.7500,7.6250,7.7500,17000,0
ABF,950309,23.0000,23.0625,22.2500,22.2500,34500,0
ABK,950309,39.7500,39.7500,38.5000,38.5000,88000,0
ABM,950309,22.6250,22.7500,22.6250,22.6250,6900,0
ABP,950309,24.5000,24.8750,24.5000,24.8750,3000,0</x-html>From ???@??? Tue Sep 23 07:20:31 1997
Received: from smtp1.nwnexus.com (smtp1.nwnexus.com [198.137.231.16])
	by mail1.halcyon.com (8.8.5/8.8.5) with ESMTP id AAA21154
	for <neal@xxxxxxxxxxxxxxxx>; Tue, 23 Sep 1997 00:28:17 -0700 (PDT)
Received: from mail.equis.com (equis.com [204.246.137.2])
	by smtp1.nwnexus.com (8.8.5/8.8.5) with ESMTP id AAA12436
	for <neal@xxxxxxxxxxx>; Tue, 23 Sep 1997 00:28:14 -0700
Received: (from list@xxxxxxxxx) by mail.equis.com (8.7.4/8.7.3) id BAA21101; Tue, 23 Sep 1997 01:16:51 -0600
Resent-Date: Tue, 23 Sep 1997 01:16:51 -0600
Message-ID: <01BCC7B5.EBA81340.chipa@xxxxxxxxxxxxxxx>
From: Chip Anderson <chipa@xxxxxxxxxxxxxxx>
To: "metastock-list@xxxxxxxxxxxxx" <metastock-list@xxxxxxxxxxxxx>
Subject: Treasury Indice Correlation Spikes?
Date: Tue, 23 Sep 1997 00:16:15 -0700
X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4211
Encoding: 36 TEXT
Resent-Message-ID: <"UOVDw2.0.d95.Yps9q"@mail.equis.com>
Resent-From: metastock-list@xxxxxxxxxxxxx
X-Mailing-List: <metastock-list@xxxxxxxxxxxxx> archive/latest/4603
X-Loop: metastock-list@xxxxxxxxxxxxx
Precedence: list
Resent-Sender: metastock-list-request@xxxxxxxxxxxxx
Status:  O

I'm easing into this stuff very slowly (and having lots of fun learning 
things in the process).  This list has been very helpful so far.  Thanks!

Tonight, I thought I'd explore some of the "fundamental truths" (pun 
intended ;-) of the market.  So I ploted the 5, 10, and 30 Year Treasury 
Bond Yield indices and compared them to Treasury Bond prices.  I used 
Metastock's correlation indicator and (surprise!) they are almost perfectly 
negatively correlated!

(Hey, I'm new at this and it was exciting to me...;-)

Just for fun, I then correlated the 3 bond indices to each other (using 
20-period MAs to smooth things).  The results surprised/confused me.  For 
much of the time, they were positively correlated above .95 as expected. 
 However, at the start of '95 and '96, there were huge spikes as the 
correlations dropped to between +.30 and +.60 respectively.

You can see what I mean by looking at http://www.nwlink.com/~chipa/TYS.htm 
 It is really quite dramatic.

So why the spikes? They seem to start appearing just before long-term trend 
changes and near the start of the year. Am I seeing things? Why have none 
occurred in 1997 yet? What would it mean if one reappeared? Can someone 
back-test this before 1994 for me (Reuters doesn't have bond index data 
before then... sigh).

If I had to guess, I'd say the the 30-year bond index turns sooner than the 
10 and 5 year indices whenever there is a major trend reversal.  Is this 
true?

Thanks for any insight,
Chip

P.S. Why doesn't Metastock's "Save As HTML" feature save color GIFs rather 
than B/W JPEGs?  Seems wrong to me. Because of that, I had to create this 
web page by hand. :-(