tadd general histograms with quartiles - cosmo - front and backend for Markov-Chain Monte Carlo inversion of cosmogenic nuclide concentrations
 (HTM) git clone git://src.adamsgaard.dk/cosmo
 (DIR) Log
 (DIR) Files
 (DIR) Refs
 (DIR) README
 (DIR) LICENSE
       ---
 (DIR) commit dab7ed7cd53163e444bc2f8d5310ac1b4a2b36d3
 (DIR) parent 6728f85140437a9c1e2020adbc0d5899e748655d
 (HTM) Author: Anders Damsgaard <anders.damsgaard@geo.au.dk>
       Date:   Mon,  9 Nov 2015 13:56:15 +0100
       
       add general histograms with quartiles
       
       Diffstat:
         M index.php                           |      40 +++++++++++++++++++++++++++++++
         M matlab/generate_plots.m             |      20 +++++++++++++++++---
       
       2 files changed, 57 insertions(+), 3 deletions(-)
       ---
 (DIR) diff --git a/index.php b/index.php
       t@@ -76,6 +76,46 @@ if (isset($_GET['wait_id']) && !empty($_GET['wait_id'])) {
                        <div class="card">
                          <div class="card-image">
                          <img src="output/<?php
       +                        echo($_GET['results_id']); ?>_Walks-7.png">
       +                        <span class="card-title blue-text text-darken-2">
       +                            Generalized model parameter values</span>
       +                  </div>
       +                  <div class="card-content">
       +                      <p>The histograms show the distribution of (a)
       +                      interglacial erosion rate, (b) glacial erosion rate, (c)
       +                      timing of last deglaciation, and (d)
       +                      d<sup>18</sup>O<sub>threshold</sub> levels that provide
       +                      the best fit to the supplied TCN concentrations. 
       +                      The vertical axis indicates the number of
       +                      simulations included in each bin out of the 10,000
       +                      simulations that followed the MCMC burn-in phase from each
       +                      MCMC walker.  The solid magenta lines denote the median
       +                      values (second quartile), while the dashed magenta lines
       +                      denote the lower and upper quartiles (25th and 75th
       +                      percentiles, respectively).</p>
       +                  </div>
       +                  <div class="card-action">
       +                  <a href="output/<?php
       +                        echo($_GET['results_id']); ?>_Walks-7.png"
       +                    target="_blank">Link to PNG</a>
       +                  <a href="output/<?php
       +                        echo($_GET['results_id']); ?>_Walks-7.pdf"
       +                    target="_blank">Link to PDF</a>
       +                  <a href="output/<?php
       +                        echo($_GET['results_id']); ?>_Walks-7.fig"
       +                    target="_blank">Link to FIG</a>
       +                  </div>
       +                </div>
       +              </div>
       +            </div>
       +
       +
       +            <div class="row">
       +                <!--<div class="col s12 m8 offset-m2">-->
       +              <div class="col s12 m10 offset-m1">
       +                <div class="card">
       +                  <div class="card-image">
       +                  <img src="output/<?php
                                echo($_GET['results_id']); ?>_Walks-5.png">
                                <span class="card-title blue-text text-darken-2">
                                    Model parameter values per inversion walker</span>
 (DIR) diff --git a/matlab/generate_plots.m b/matlab/generate_plots.m
       t@@ -210,7 +210,7 @@ for i1 = 1:M % for each model parameter
            subplot(M,Nwalkers,isub)
            %Nhistc=histc(Ss{iwalk}.ms(i1,:),xbins{i1});
            %bar(xbins{i1},Nhistc,'histc')
       -    histogram(Ss{iwalk}.ms(i1,:), xbins{i1}, 'Normalization', 'probability');
       +    histogram(Ss{iwalk}.ms(i1,:), xbins{i1});
            
            if i1 == 1
                title(['MCMC walker ' num2str(iwalk)])
       t@@ -329,12 +329,26 @@ for i1 = 1:M % for each model parameter
            for iwalker=1:Nwalkers
                data = [data, Ss{iwalker}.ms(i1,:)];
            end
       -    Nhistc=histc(data, xbins{i1});
       -    bar(xbins{i1},Nhistc,'histc')
       +    
       +    hold on
       +    %Nhistc=histc(data, xbins{i1});
       +    %bar(xbins{i1},Nhistc,'histc')
       +    histogram(data, xbins{i1});
        
       +    % 2nd quartile = median = 50th percentile
            med = median(data);
            plot([med, med], get(gca,'YLim'), 'm-')
            
       +    % 1st quartile = 25th percentile
       +    prctile25 = prctile(data, 25);
       +    plot([prctile25, prctile25], get(gca,'YLim'), 'm--')
       +    
       +    % 3rd quartile = 75th percentile
       +    prctile75 = prctile(data, 75);
       +    plot([prctile75, prctile75], get(gca,'YLim'), 'm--')
       +    
       +    hold off
       +    
            if i1 == 1
                xlabel('Interglacial erosion rate [mm/yr]')
                text(0.02,0.98,'a', 'Units', ...