Mathematics – Probability
Scientific paper
Jan 2010
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2010aas...21537008c&link_type=abstract
American Astronomical Society, AAS Meeting #215, #370.08; Bulletin of the American Astronomical Society, Vol. 42, p.562
Mathematics
Probability
Scientific paper
In the statistical analysis of survey data, a large number of data points having a continuously distributed observed variable may be grouped into ranges of constant width, a process known as "binning". For example, galaxies are often placed into groups by redshift. In binning data, a certain amount of information about the object is often lost: any information at a higher degree of accuracy than needed to place it into a bin is discarded. A methodology is proposed for the determination of population distributions allowing for full retention of the measured value for each observation in cases where the uncertainties are expected to be Gaussian. If the uncertainties are normally distributed, a distinct Gaussian function may be determined for each measurement, with the observed value as the mean and the uncertainty as the standard deviation. These Gaussian distributions - one for each observation - are then summed, creating a continuous probability density distribution. In this manner, all available data may be incorporated into the final population distribution without loss of information due to binning. Because the contribution of each individual point to the overall distribution falls off exponentially with distance from the point, small variations from a true Gaussian for the errors will not significantly impact the final overall distribution. Therefore, this technique may be applied for error distributions that are only approximately Gaussian. This research has been supported in part by the Scott Smith Research Fund at the University of Toledo Department of Physics and Astronomy.
No associations
LandOfFree
Continuous Probability Density Distribution as an Alternative to Binning in the Statistical Analysis of Survey Data does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Continuous Probability Density Distribution as an Alternative to Binning in the Statistical Analysis of Survey Data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Continuous Probability Density Distribution as an Alternative to Binning in the Statistical Analysis of Survey Data will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-963968