Table 3
Comparison of the performances of our MLPQNA (here labeled as QNA) method against all other empirical methods analyzed by PHAT board.
| A | 18-band; |Δz| ≤ 0.15 | 14-band; |Δz| ≤ 0.15 | 18-band; R < 24; |Δz| ≤ 0.15 | 14-band; R < 24; |Δz| ≤ 0.15 | ||||||||
|
|
||||||||||||
| Code | bias | scatter | outliers % | bias | scatter | outliers % | bias | scatter | outliers % | bias | scatter | outliers % |
|
|
||||||||||||
| QNA | 0.0006 | 0.056 | 16.3 | 0.0028 | 0.063 | 19.3 | 0.0002 | 0.053 | 11.7 | 0.0016 | 0.060 | 13.7 |
| AN-e | –0.010 | 0.074 | 31.0 | –0.006 | 0.078 | 38.5 | –0.013 | 0.071 | 24.4 | –0.007 | 0.076 | 32.8 |
| EC-e | –0.001 | 0.067 | 18.4 | 0.002 | 0.066 | 16.7 | –0.006 | 0.064 | 14.5 | –0.003 | 0.064 | 13.5 |
| PO-e | –0.009 | 0.052 | 18.0 | –0.007 | 0.051 | 13.7 | –0.009 | 0.047 | 10.7 | –0.008 | 0.046 | 7.1 |
| RT-e | –0.009 | 0.066 | 21.4 | –0.008 | 0.067 | 24.2 | –0.012 | 0.063 | 16.4 | –0.012 | 0.064 | 18.4 |
|
|
||||||||||||
| B | 18-band; |Δz| ≤ 0.5 | 14-band; |Δz| ≤ 0.5 | 18-band; R < 24; |Δz| ≤ 0.5 | 14-band; R < 24; |Δz| ≤ 0.5 | ||||||||
|
|
||||||||||||
| Code | bias | scatter | outliers % | bias | scatter | outliers % | bias | scatter | outliers % | bias | scatter | outliers % |
|
|
||||||||||||
| QNA | –0.0028 | 0.114 | 3.8 | –0.0046 | 0.125 | 3.8 | –0.0039 | 0.101 | 1.7 | –0.0039 | 0.101 | 1.7 |
| AN-e | –0.036 | 0.151 | 3.1 | –0.035 | 0.173 | 4.2 | –0.047 | 0.130 | 1.4 | –0.047 | 0.130 | 1.4 |
| EC-e | –0.007 | 0.120 | 3.6 | –0.003 | 0.114 | 3.6 | –0.015 | 0.106 | 1.9 | –0.015 | 0.106 | 1.9 |
| PO-e | –0.013 | 0.124 | 3.1 | 0.001 | 0.107 | 2.3 | –0.020 | 0.098 | 1.2 | –0.020 | 0.098 | 1.2 |
| RT-e | –0.031 | 0.126 | 3.2 | –0.028 | 0.137 | 3.6 | –0.034 | 0.111 | 1.4 | –0.034 | 0.111 | 1.4 |
|
|
||||||||||||
| C | 18-band; zsp ≤ 1.5, |Δz| ≤ 0.15 | 14-band; zsp ≤ 1.5, |Δz| ≤ 0.15 | 18-band; zsp > 1.5, |Δz| ≤ 0.15 | 14-band; zsp > 1.5, |Δz| ≤ 0.15 | ||||||||
|
|
||||||||||||
| Code | bias | scatter | outliers % | bias | scatter | outliers % | bias | scatter | outliers % | bias | scatter | outliers % |
|
|
||||||||||||
| QNA | –0.0004 | 0.053 | 14.6 | 0.0001 | 0.061 | 16.6 | 0.0074 | 0.072 | 26.3 | 0.0222 | 0.070 | 35.0 |
| AN-e | –0.017 | 0.070 | 27.6 | –0.010 | 0.076 | 33.6 | 0.051 | 0.078 | 50.7 | 0.045 | 0.077 | 66.4 |
| EC-e | –0.003 | 0.065 | 16.1 | –0.000 | 0.064 | 14.5 | 0.015 | 0.077 | 32.3 | 0.015 | 0.077 | 29.5 |
| PO-e | –0.012 | 0.049 | 12.6 | –0.011 | 0.047 | 9.4 | 0.019 | 0.075 | 48.3 | 0.026 | 0.074 | 37.7 |
| RT-e | –0.016 | 0.062 | 19.6 | –0.014 | 0.064 | 21.1 | 0.040 | 0.072 | 31.8 | 0.039 | 0.071 | 41.9 |
Notes. For a description of other methods (namely AN-e, EC-e, PO-e and RT-e) see the text. The table is divided into three parts (namely A, B and C). Data for the other empirical method have been extracted from Hildebrandt et al. (2010). In each part of the table we list the results (on both the 18 and the 14 bands datasets) for a specific subsample of the PHAT objects. Part A: statistical indicators (bias and scatter) for the 18 and 14 bands computed on objects with |Δz| ≤ 0.15 and for objects with |Δz| ≤ 0.15 and R < 24. The column “outliers” gives the fraction of outliers defined as objects with |Δz| > 0.15. Part B: the same but for |Δz| ≤ 0.5. Part C: the same but for objects with spectroscopic redshift zsp ≤ 1.5 and |Δz| ≤ 1.5, and for zsp > 1.5 and |Δz| ≤ 1.5. The definitions of bias, scatter, and outliers fraction are given in the text. Values were computed by the PHAT collaboration on the whole PHAT1 dataset.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.