The First Photometric Study of AH Mic Contact Binary System
Authors:
Atila Poro,
Mark G. Blackford,
Selda Ranjbar Salehian,
Esfandiar Jahangiri,
Meysam Samiei Dastjerdi,
Mohammadjavad Gozarandi,
Reihaneh Karimi,
Tabassom Madayen,
Elnaz Bakhshi,
Farnad Hedayati
Abstract:
The first multi-color light curve analysis of the AH Mic binary system is presented. This system has very few past observations from the southern hemisphere. We extracted the minima times from the light curves based on the Markov Chain Monte Carlo (MCMC) approach and obtained a new ephemeris. To provide modern photometric light curve solutions, we used the Physics of Eclipsing Binaries (Phoebe) so…
▽ More
The first multi-color light curve analysis of the AH Mic binary system is presented. This system has very few past observations from the southern hemisphere. We extracted the minima times from the light curves based on the Markov Chain Monte Carlo (MCMC) approach and obtained a new ephemeris. To provide modern photometric light curve solutions, we used the Physics of Eclipsing Binaries (Phoebe) software package and the MCMC approach. Light curve solutions yielded a system temperature ratio of 0.950, and we assumed a cold star-spot for the hotter star based on the O'Connell effect. This analysis reveals that AH Mic is a W-subtype W UMa contact system with a fill-out factor of 21.3% and a mass ratio of 2.32. The absolute physical parameters of the components are estimated by using the Gaia Early Data Release 3 (EDR3) parallax method to be M_h(M_Sun)=0.702(26), M_c(M_Sun)=1.629(104), R_h(R_Sun)=0.852(21), R_c(R_Sun)=1.240(28), L_h(L_Sun)=0.618(3) and L_c(L_Sun)=1.067(7). The orbital angular momentum of the AH Mic binary system was found to be 51.866(35). The components' positions of this system are plotted in the Hertzsprung-Russell (H-R) diagram.
△ Less
Submitted 5 April, 2022; v1 submitted 2 April, 2022;
originally announced April 2022.
Horizon-Independent Optimal Prediction with Log-Loss in Exponential Families
Authors:
Peter Bartlett,
Peter Grunwald,
Peter Harremoes,
Fares Hedayati,
Wojciech Kotlowski
Abstract:
We study online learning under logarithmic loss with regular parametric models. Hedayati and Bartlett (2012b) showed that a Bayesian prediction strategy with Jeffreys prior and sequential normalized maximum likelihood (SNML) coincide and are optimal if and only if the latter is exchangeable, and if and only if the optimal strategy can be calculated without knowing the time horizon in advance. They…
▽ More
We study online learning under logarithmic loss with regular parametric models. Hedayati and Bartlett (2012b) showed that a Bayesian prediction strategy with Jeffreys prior and sequential normalized maximum likelihood (SNML) coincide and are optimal if and only if the latter is exchangeable, and if and only if the optimal strategy can be calculated without knowing the time horizon in advance. They put forward the question what families have exchangeable SNML strategies. This paper fully answers this open problem for one-dimensional exponential families. The exchangeability can happen only for three classes of natural exponential family distributions, namely the Gaussian, Gamma, and the Tweedie exponential family of order 3/2. Keywords: SNML Exchangeability, Exponential Family, Online Learning, Logarithmic Loss, Bayesian Strategy, Jeffreys Prior, Fisher Information1
△ Less
Submitted 19 May, 2013;
originally announced May 2013.
