Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userImplementation of a goodness-of-fit test for finding optimal concurrent radio and {gamma}-ray pulsar light curves
329
0
0.0
(
0
)
Added by
Albertus Seyffert mr
Publication date
2016
fields
Physics
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
Since the launch of the Fermi Large Area Telescope in 2008 the number of known ${gamma}$-ray pulsars has increased immensely to over 200, many of which are also visible in the radio and X-ray bands. Seyffert et al. (2011) demonstrated how constraints on the viewing geometries of some of these pulsars could be obtained by comparing their observed radio and ${gamma}$-ray light curves by eye to light curves from geometric models. While these constraints compare reasonably well with those yielded by more rigorous single-wavelength approaches, they are still a somewhat subjective representation of how well the models reproduce the observed radio and ${gamma}$-ray light curves. Constructing a more rigorous approach is, however, made difficult by the large uncertainties associated with the ${gamma}$-ray light curves as compared to those associated with the radio light curves. Naively applying a ${chi}^{2}$-like goodness-of-fit test to both bands invariably results in constraints dictated by the radio light curves. A number of approaches have been proposed to address this issue. In this paper we investigate these approaches and evaluate the results they yield. Based on what we learn, we implement our own version of a goodness-of-fit test, which we then use to investigate the behaviour of the geometric models in multi-dimensional phase space.
rate research
Read More
Guillemot et al. recently reported the discovery of $gamma$-ray pulsations from the 22.7ms pulsar (pulsar A) in the famous double pulsar system J0737-3039A/B. The $gamma$-ray light curve (LC) of pulsar A has two peaks separated by approximately half a rotation, and these are non-coincident with the observed radio and X-ray peaks. This suggests that the $gamma$-ray emission originates in a part of the magnetosphere distinct from where the radio and X-ray radiation is generated. Thus far, three different methods have been applied to constrain the viewing geometry of pulsar A (its inclination and observer angles $alpha$ and $zeta$): geometric modelling of the radio and $gamma$-ray light curves, modelling of the position angle sweep in phase seen in the radio polarisation data, and independent studies of the time evolution of the radio pulse profile of pulsar A. These three independent, complementary methods have yielded consistent results: pulsar As rotation axis is likely perpendicular to the orbital plane of the binary system, and its magnetic axis close to lying in the orbital plane (making this pulsar an orthogonal rotator). The observer is furthermore observing emission close to the magnetic axis. Thus far, however, current models could not reproduce all the characteristics of the radio and $gamma$-ray light curves, specifically the large radio-to-$gamma$ phase lag. In this paper we discuss some preliminary modelling attempts to address this problem, and offer ideas on how the LC fits may be improved by adapting the standard geometric models in order to reproduce the profile positions more accurately.
We previously obtained constraints on the viewing geometries of 6 Fermi LAT pulsars using a multiwavelength approach (Seyffert et al., 2011). To obtain these constraints we compared the observed radio and $gamma$-ray light curves (LCs) for those 6 pulsars by eye to LCs predicted by geometric models detailing the location and extent of emission regions in a pulsar magnetosphere. As a precursor to obtaining these constraints, a parameter study was conducted to reinforce our qualitative understanding of how the underlying model parameters effect the LCs produced by the geometric models. Extracting useful trends from the $gamma$-ray model LCs proved difficult though due to the increased complexity of the geometric models for the $gamma$-ray emission relative to those for the radio emission. In this paper we explore a second approach to investigating the interplay between the model parameters and the LC atlas. This approach does not attempt to understand how the set of model parameters influences the LC shapes directly, but rather, more fundamentally, investigates how the set of model parameters effects the sky maps from which the latter are extracted. This allows us to also recognise structure within the atlas itself, as we are now able to attribute certain features of the LCs to specific features on the sky map, meaning that we not only understand how the structure of single LCs come about, but also how their structure changes as we move through the geometric solution space.
Statistical modeling plays a fundamental role in understanding the underlying mechanism of massive data (statistical inference) and predicting the future (statistical prediction). Although all models are wrong, researchers try their best to make some of them be useful. The question here is how can we measure the usefulness of a statistical model for the data in hand? This is key to statistical prediction. The important statistical problem of testing whether the observations follow the proposed statistical model has only attracted relatively few attentions. In this paper, we proposed a new framework for this problem through building its connection with two-sample distribution comparison. The proposed method can be applied to evaluate a wide range of models. Examples are given to show the performance of the proposed method.
In many fields, data appears in the form of direction (unit vector) and usual statistical procedures are not applicable to such directional data. In this study, we propose non-parametric goodness-of-fit testing procedures for general directional distributions based on kernel Stein discrepancy. Our method is based on Steins operator on spheres, which is derived by using Stokes theorem. Notably, the proposed method is applicable to distributions with an intractable normalization constant, which commonly appear in directional statistics. Experimental results demonstrate that the proposed methods control type-I error well and have larger power than existing tests, including the test based on the maximum mean discrepancy.
Recently there have been many research efforts in developing generative models for self-exciting point processes, partly due to their broad applicability for real-world applications. However, rarely can we quantify how well the generative model captures the nature or ground-truth since it is usually unknown. The challenge typically lies in the fact that the generative models typically provide, at most, good approximations to the ground-truth (e.g., through the rich representative power of neural networks), but they cannot be precisely the ground-truth. We thus cannot use the classic goodness-of-fit (GOF) test framework to evaluate their performance. In this paper, we develop a GOF test for generative models of self-exciting processes by making a new connection to this problem with the classical statistical theory of Quasi-maximum-likelihood estimator (QMLE). We present a non-parametric self-normalizing statistic for the GOF test: the Generalized Score (GS) statistics, and explicitly capture the model misspecification when establishing the asymptotic distribution of the GS statistic. Numerical simulation and real-data experiments validate our theory and demonstrate the proposed GS tests good performance.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
