Re: mathematical problem, adjusting readings

Wow, the guy from the great mindprod website I always end up at when researching a Java thing. ;)

Anyway, I work in the data acquisition industry but is in no way an expert, yet your problem sounds very familiar so here are my 2 cents should you be interested.

A bell curve (2'nd degree polynomial) is less crude than linear interpolation but still real world data is anything but smooth and symmetric. An n'th degree polynomial could be calculated with a bunch of data points you provide. Norwegian GeoSoft has some nice utilities for this based on Java Numerics:
http://geosoft.no/software/index.html

Now inverse this curve (bell upside down) and add your datapoint, the normalized value at any given time can now be found by simply taking an average of the Y-value of your datapoint and the Y-value of the inverse curve.

Anyway that's how I would approach the problem, unless I misunderstood something. Thanks for your help on many occasions.

/Casper




Roedy Green wrote:
I probably could solve this myself, but I thought I would share this
problem with the group because you might find it interesting or mildly
challenging.

The problem is this. I sample the hit counts on my website each day,
ideally at noon. I then graph them. I also graph a moving average.
The problem is I am often late or early in my daily sampling, or the
website might be down, so I can't do it at noon.

When I sample late it generates the visual impression of a very good
day followed by a very bad day, even though nothing unusual actually
happened.

So here is the problem. How do I adjust the raw figures to get the
estimated figures for what would have been the count as of noon?

1. cut 1. Use linear interpolation.

2. cut 2. Account for the average distribution -- i.e. some times of
day are busier that others. I could create this by hourly sampling
over a week to discover the shape of the distribution curve.

Let's say for example it was a bell shaped curve around noon. I could
provide you with a bell shaped curve function, that you would in some
mysterious way use to adjust the daily figure. Just how is this curve
normalised?


.

Relevant Pages

  • Re: Lightwaver to Rhino
    ... Your stuff looks very impressive Jeff. ... I just counted and there are 86 parts in this particular valve assembly. ... BTW - I did have a look around at your website and your work is excellent. ... create a curve that can then be extruded, made into a planar face, ...
    (comp.graphics.apps.lightwave)
  • Re: God=G_uv proves 40k B.C. Creation
    ... both graphs in Hammond's diagram ... the "brain growth deficit" (BGD) ... curve at any point. ... website is wrong too. ...
    (sci.physics.relativity)
  • Bell shaped curves
    ... not all data forming a symmetrical bell shaped curve are normal. ... from 15 to 265 and their frequencies have a peak value ... for the total set ...
    (sci.math.num-analysis)
  • Re: God=G_uv proves 40k B.C. Creation
    ... in your website: ... the "brain growth deficit" (BGD) ... curve at any point. ...
    (sci.physics.relativity)
  • Re: God=G_uv proves 40k B.C. Creation
    ... > your website. ... > start at zero with no visible difference at the zero point. ... > curve at any point. ... I thought your point was the delta between the two curves - which is BGD ...
    (sci.physics.relativity)