I have had a similar request from my son who wanted to isolate "significant"
peaks in data (coming from biological experiments). I gave him a helping hand
in doing the preprocessing that removes the variations that are "not
significant". It did, however, some time to explain this, but he got it,
Brian,
Jonathan, Jerry, et al
As to the number of potential observations. Thousands of series,
10s to hundreds of thousands of observations per series at a
minimum. Possibly millions. So computational efficiency isn't a
luxury, it is a requirement.
Brian
The following method is fairly
p with some matrix manipulations - when I get
>> to my computer I will explain further.
>>
>>
>> Sent from my Verizon Wireless BlackBerry
>>
>> -Original Message-
>> From: Bryan McCormick
>> Sender: use-livecode-boun...@lists.runrev.com
>
xplain further.
>
>
> Sent from my Verizon Wireless BlackBerry
>
> -Original Message-
> From: Bryan McCormick
> Sender: use-livecode-boun...@lists.runrev.com
> Date: Sun, 05 Dec 2010 02:12:29
> To:
> Reply-To: How to use LiveCode
> Subject: Re: Finding lo
-To: How to use LiveCode
Subject: Re: Finding local minima and maxima of a graph
Jonathan, Jerry, et al
As to the number of potential observations. Thousands of series, 10s to
hundreds of thousands of observations per series at a minimum. Possibly
millions. So computational efficiency isn
Jonathan, Jerry, et al
As to the number of potential observations. Thousands of series, 10s to
hundreds of thousands of observations per series at a minimum. Possibly
millions. So computational efficiency isn't a luxury, it is a requirement.
Yes, absolutely, detrending is required. I won't go
Message: 26
Date: Sat, 04 Dec 2010 12:56:05 -0500
From: Bryan McCormick
To: use-livecode@lists.runrev.com
Subject: Finding local minima and maxima of a graph
Message-ID: <4cfa80b5.3090...@deepfoo.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
I have a very large number of grap
local minima and maxima of a graph
Yes, what Jonathan wrote.
The folks who do heavy computing in the audio world do something called
windowing. Thats like a rolling average, except weighted toward the "center".
Choosing the width of the window and the algorithm for determining its
local minima and maxima of a graph
Sent: Dec 5, 2010 12:33 AM
Jonathan,
Sure, that approach can work. But it isn't terribly efficient. It would
however be less computationally intensive if I detrend first. That's the
first step. So if you have a notion on a least squares linear fit l
Yes, what Jonathan wrote.
The folks who do heavy computing in the audio world do something called
windowing. Thats like a rolling average, except weighted toward the "center".
Choosing the width of the window and the algorithm for determining its shape
can get complicated, and has been greatly
Jonathan,
Sure, that approach can work. But it isn't terribly efficient. It would
however be less computationally intensive if I detrend first. That's the
first step. So if you have a notion on a least squares linear fit line,
that would be most welcome.
_
Hi Bryan,
The first step is to define a local minimum or maximum for this situation.
Plotting a 3-day rolling average would be pretty straight forward. Is three
days sufficient?
You could get a 5-day rolling average, then pick five day averages that are
are greater than, or less than, their neig
Jerry,
Yes, that would be far too many hits and there has to be some scaling
applied as a filter for that reason. The criteria more precisely is
"significant" peak and trough values. With significant being the trick.
For example, is the peak or trough 3 percent or more away.
There is a seaso
On Dec 4, 2010, at 7:38 PM, Bryan McCormick wrote:
> Peter, Jonathan
>
> I had thought to use a sin function on the time series to find peak and
> trough values. I would filter the sin function results for those close to 1
> or -1, though I am not sure what the filter interval will be just yet.
Peter, Jonathan
I had thought to use a sin function on the time series to find peak and
trough values. I would filter the sin function results for those close
to 1 or -1, though I am not sure what the filter interval will be just yet.
Then, I would be able to grab the value from the actual ti
Jonathan,
No, these are data files in .csv format. Column data with high and low
values from the series. Unfortunately I have thousands of these to wade
through, with this being an ongoing project. First task is to deal with
the history.
___
use-li
: Sat, 04 Dec 2010 14:10:26
To:
Reply-To: How to use LiveCode
Subject: Re: Finding local minima and maxima of a graph
Peter,
Yes, I have thousands of .csv files to process. I have data that is
ordered by column containing high and low values.
On 12/4/2010 12:56 PM, Bryan McCormick wrote:
>
Peter,
Yes, I have thousands of .csv files to process. I have data that is
ordered by column containing high and low values.
On 12/4/2010 12:56 PM, Bryan McCormick wrote:
I have a very large number of graphs to crunch through to find local
minima and maxima. The data is regularly spaced, whic
Do you have access to the data the graphs are based on? Or are you
talking about trying to analyze the shape of a curve (bitmapped or
otherwise)? The two problems will have quite different approaches.
-- Peter
Peter M. Brigham
pmb...@gmail.com
http://home.comcast.net/~pmbrig
On Dec 4, 2010
19 matches
Mail list logo
