Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>If you have the <a href="http://www.mathworks.com/products/statistics/" rel="nofollow">statistical toolbox</a>, try something like this:</p> <pre><code>a = dlmread('~\downloads\-data-anonfiles-1383150325725.txt'); % read data p = mvnpdf(a,mean(a),cov(a)); % multivariate PDF of your data p_sample = numel(p)*p/sum(p); % normalize pdf to number of samples thresh = 0.5; % set an arbitrary threshold to filter idx_thresh = p_sample &gt; thresh; % logical indices of samples that meet the threshold a_filtered = a(idx_thresh,:); </code></pre> <p>then repeat this again with filtered data.</p> <pre><code> p = mvnpdf(a,mean(a_filtered),cov(a_filtered)); p_sample = numel(p)*p/sum(p); % normalize pdf to number of samples thresh = 0.1; % set an arbitrary threshold to filter idx_thresh = p_sample &gt; thresh; % logical indices of samples that meet the threshold a_filtered = a_filtered (idx_thresh,:); </code></pre> <p>I was able to pull out most of the dominant distribution in just 2 iterations. But I think you would want to repeat until the mean(a_filtered) and cov(a_filtered) reached steady state values. Plot them as a function of iteration, and when they approach a flat line then you've found the correct values.</p> <p>This is equivalent to filtering with an rotated ellipse, but IMO it is easier and more useful because now you actually have the 5 <a href="http://www.mathworks.com/help/stats/multivariate-normal-distribution.html" rel="nofollow">mvnpdf</a> parameters (mu_x, mu_y, sigma_xx, sigma_yy, sigma_xy) required to reproduce the distribution. If you model the iso-line (p(x,y) = thresh) as an rotated ellipse, you would have to manipulate the minor and major axes (a,b), the translation coordinates (h,k) and the rotation (theta) to get the mvnpdf parameters.</p> <p>Then after extracting the first distribution, you can repeat the process to find the secondary distribution.</p>
    singulars
    1. This table or related slice is empty.
    1. POEliminating all the 2D datapoints outside an ellipse
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USMark Mikofski
    1. USMark Mikofski
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. POEliminating all the 2D datapoints outside an ellipse
      singulars
      1. PTQuestion
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTAcceptedByOriginator
    1. CONice solution! But actually, the isolines of a twodimensional normal distribution are ellipses. And both the distribution and the ellipse are described by five parameters (the covariance matrix is symmetric).
      singulars
      1. PO
      1. USA. Donda
    2. CO@A.Donda thanks, I updated answer based on your comment
      singulars
      1. PO
      1. USMark Mikofski
    3. COFiltering out cells with x-axis values less than 0.65e5 and reducing the threshold to 0.3, does a suitable job on the first try. Thanks very much @MarkMikofski for the solution
      singulars
      1. PO
      1. USLee Sande
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload