Hello Gary,

Concerning QP mutual fund data, I created a spreadsheet that 
identifies for each ticker, the % price change day over day, and the 
date it occurs, when this % is greater than or equal to 5%...this is 
a 1 day percentage change. As you will see, there are many days here 
the change is 10%, 25%, 100% or more. A few have $ change over 1000%.
The spreadsheet sorts these % changes from highest to lowest. There 
is a tab that has all the data on it..and a separate tab segregating 
it by year. Only years 1998 thru 2006 are included, because for some 
reason the scan I ran didn't go back any further (I don't know 
why..could it be memory limitation of scan program..don't know)..but 
this is enough for now.

The spreadsheet is in zipped format and is called 
MF_pctchg_ge5d.zip .
It is in the files section of this group. After you download and 
copy it, you of course, can delete it so it doesn't take up space on 
the system. The file was too large to send in one of these group 
messages.

Please note, that I am only doing this so that we may all work 
together to clean up the data. Please note that in many cases, it is 
not possible to do 'due diligence' and examine a graph for errors 
before we make a purchase. I've noticed that many times, the error 
is a single data point that simply is not visible to the naked eye , 
at least not easily..and certainly if you aren't looking for and 
know where to look. This is truer the further back the error is. Yet 
some of these aberrations are HUGE!..and need to be fixed. As 
someone else suggested, these are not simply errors due to not yet 
recording capital gains/divvy's that have been paid this month. 
These errors occurr uniformly across the years studied. Some are 
very obvious...like a single datapoint 35$ or so higher or lower
than the price before or after. Many of these points, as I've said , 
are NOT visible to the eye...yet can really mess up a computer 
analysis.

As time permits, I will try to develop other scans/tests to clean up 
data issues. One other I've notices, that is not too common, is an 
issue will have data for years, then go years with blanks, then have 
data again at a completely different price level. I mentioned an 
example of this in an earlier post, but don't recall the ticker.

I spent alot of time on this, and hope you look at it. Thx.

Investor0329

Reply via email to