Changes between Version 24 and Version 25 of Stack_Sky_Levels

Timestamp:

Sep 19, 2014, 4:31:08 PM (12 years ago)

Author:

watersc1

Comment:

--

Stack_Sky_Levels

@@ -28 +28 @@
 || stack || -0.0428245 || 228.501 ||
 
-These values can be used to check that the psphot claimed background level is a function of the sky sigma, as the bias report below discusses.  Plotted are three realizations of each warp: from my ppStack process (1), from Peter's ppStack process (3), and I thought that was another psphot run, but it's Peter doing what I claimed I just did but getting a different answer (2).  Biased because I clipped at +/- 100?  I'll need to fix this bit.
-
-[[Image(sigma_bias.png,200px)]]
-
-Looking at the difference between psphot and the full image fits (to confirm that the outlier does have the same bias as the other images):
-
-[[Image(sigma_bias_cor.png,200px)]]
+These values can be used to check that the psphot claimed background level is a function of the sky sigma, as the bias report from May below discusses.  Plotted are three realizations of each warp's: (1) is my ppStack run, (3) is Peter's ppStack run, and I originally thought (2) was also a psphot background level, but reading Peter's page closer, this is actually his manual histogram/Gaussian fit data.  Due to the quantized nature of the warp pixel data, differences in histogram bin size can influence the fit results.  
+
+[[Image(sigma_bias.png,400px)]]
+
+Subtracting off my "true mean" values corrects the outlier warp to fall closer to the same trend as the other points at smaller sigma.  My understanding of this remaining trend is that even when I believe the warp pixel value mean to be approximately zero, psphot still produces a measured background that is inflated slightly, with an increase of about 0.3 counts.
+
+[[Image(sigma_bias_cor.png,400px)]]
+
+On the stacks, we seem to be getting a better estimate of the true sky level, but there is still some variation depending on which set of pixels is examined.  Below I show the CMF sky values for multiple photometry runs with psphot and psphotStack that I ran with my stack.  The mean of this distribution (which samples the background model realization) shifts by about 1 count between these runs, due to this pixel selection.  Also plotted is a fixed seed comparison between psphot and psphotStack, again confirming that they produce the same models. 
 
 [[Image(multiple_stack_runs.png)]]
+
+Finally, plotting all the warp and stack background models as a function of pixel index (index 0 is image(0,0), and index 256 is image(16,16), it is clear that the warp background models all cluster around values ~0.3.  The stack matches the expected value around zero.  
 
 [[Image(warp_stack_pixel_comp.png)]]