We investigated the statistical properties of both cool and hot starspots in eclipsing binary stars. The starspot and binary parameters for contact and semi-detached systems were collected from literature, which were determined on the basis of synthetic light-curve analysis. We examined associations between these parameters. As a result, the cool spots in W-type binaries show properties similar to those of sunspots and starspots generated by dynamos, which differs from those of the cool spots in A-type binaries. The properties of hot spots also differ between the W- and A-type samples. From the physical properties of A- and W-type binaries, we infer that mass transfer is a dominant process for forming the hot spots in A-type binaries; and both mass transfer and magnetic activity can contribute to the formation of the hot spots in W-type binaries. Our results also indicate that the hot-spot size in the A-type sample is correlated with the temperature of spotted stars, orbital period, mass ratio, and fill-out factor.

We selected mass-transferring binary candidates from the catalog of Kepler eclipsing binary stars and investigated the dependence of the mass-transfer rate on several astrophysical quantities, including orbital period, semi-major axis, mass ratio, fill-out factor, temperature, and mass. We selected these candidates using $O-C$ diagrams and calculated their mass-transfer rates. Primary masses were obtained from the mass--temperature relation, and the temperatures of the component stars were extracted from a catalog of temperatures for Kepler eclipsing binary stars. The mass-transfer rates of overcontact systems have associations with astrophysical quantities that seem to differ from those of semi-detached or detached systems. These associations indicate that mass exchange from more- to less-massive components (from less- to more-massive components) generally becomes rapid (slow) as the mass exchange evolves. However, for mass exchange from more- to less-massive components, this tendency is not reasonable for binaries with a short period ($P<0.4$ d) and low mass ($M_1<1.2$ M$_\odot$) because the correlations of these binaries are opposite to those of binaries with long period and high mass. These different correlations likely arise from differences in correlation between subtypes of W UMa systems (i.e., W- and A-types). Alternatively, when mass exchange from more- to less-massive components reoccurs after a mass-ratio reversal, its properties may differ from those of the first mass exchange.

We present the properties of the ensemble variability $V$ for nearly 5000 near-infrared (NIR) AGNs selected from the catalog of Quasars and Active Galactic Nuclei (13th Ed.) and the SDSS-DR7 quasar catalog. From 2MASS, DENIS, and UKIDSS/LAS point source catalogs, we extract 2MASS-DENIS and 2MASS-UKIDSS counterparts for cataloged AGNs by catalog cross-identification. We further select variable AGNs based on an optimal criterion for selecting the variable sources. The sample objects are divided into subsets according to whether NIR light originates by optical or NIR emission in the rest frame; and we examine the correlations of the ensemble variability with the rest-frame wavelength, redshift, luminosity, and rest-frame time lag. In addition, we also examine the correlations of variability amplitude with optical variability, radio intensity, and radio-to-optical flux ratio. The rest-frame optical variability of our samples shows known negative correlations with luminosity and positive correlations with rest-frame time lag (i.e., the structure function, SF). However, no well-known negative correlation exists between the rest wavelength and optical variability. This inconsistency might be due to a biased sampling of high-redshift AGNs. NIR variability in the rest frame is anticorrelated with the rest wavelength, which is consistent with previous suggestions. However, correlations of NIR variability with luminosity and rest-frame time lag are the opposite of these correlations of the optical variability; that is, the NIR variability is positively correlated with luminosity but negatively correlated with the rest-frame time lag. Because these trends are qualitatively consistent with the properties of radio-loud quasars reported by some previous studies, most of our sample objects are probably radio-loud quasars. Finally, we also discuss the negative correlations seen in the NIR SFs.

We report on the near-infrared selected AGN candidates extracted from 2MASS/ROSAT catalogues and discuss their properties. First, near-infrared counterparts of a X-ray source in ROSAT catalogues (namely, Bright Source Catalogue (BSC) and Faint Source Catalogue (FSC)) were extracted by positional cross-identification of <=30''. Because these counterparts would contain many mis-identifications, we further imposed near-infrared colour selection criteria and extracted reliable AGN candidates (BSC: 5,273, FSC: 10,071). Of 5,273 (10,071) candidates in the BSC (FSC), 2,053 (1,008) are known AGNs. Near-infrared and X-ray properties of candidates show similar properties with known AGNs and are consistent with previous studies. We also searched for counterparts in other wavelengths (that is, optical, near-infrared, and radio), and investigated properties in multiwavelength. No significant difference between known AGNs and unclassified sources could be seen. However, some unclassified sources in the FSC showed slightly different properties compared with known AGNs. Consequently, it is highly probable that we could extract reliable AGN candidates though candidates in the FSC might be spurious.

We propose near-infrared colour selection criteria to extract Active Galactic Nuclei (AGNs) using the near-infrared Colour-Colour Diagram (CCD) and predict near-infrared colour evolution with respect to redshift. First, we cross-identify two AGN catalogues with the 2MASS Point Source Catalogue, and confirm both the loci of quasars/AGNs in the near-infrared CCD and redshift-colour relations. In the CCD, the loci of over 70 - 80% of AGNs can be distinguished from the stellar locus. To examine the colours of quasars, we simulate near-infrared colours using Hyperz code. Assuming a realistic quasar SED, we derive simulated near-infrared colours of quasars with redshift (up to z ~ 11). The simulated colours can reproduce not only the redshift-colour relations but also the loci of quasars/AGNs in the near-infrared CCD. We finally discuss the possibility of contamination by other types of objects. We compare the locus of AGNs with the other four types of objects (namely, microquasars, CVs, LMXBs, and MYSOs) which have a radiation mechanism similar to that of AGNs. In the near-infrared CCD, each type of object is located at a position similar to the stellar locus. Accordingly, it is highly probable that the four types of objects can be distinguished on the basis of the locus in a near-infrared CCD. We additionally consider contamination by distant normal galaxies. The near-infrared colours of several types of galaxies are also simulated using the Hyperz code. Although galaxies with z ~ 1 have near-infrared colours similar to those of AGNs, these galaxies are unlikely to be detected because they are very faint. In other words, few galaxies should contaminate the locus of AGNs in the near-infrared CCD. Consequently, we can extract reliable AGN candidates on the basis of the near-infrared CCD.

We present a catalogue of variable stars in the near-infrared wavelength detected with overlapping regions of the 2MASS public images, and discuss their properties. The investigated region is in the direction of the Galactic center (-30 < l < 20, |b| < 20), which covers the entire bulge. We have detected 136 variable stars, of which 6 are already-known and 118 are distributed in |b| < 5 region. Additionally, 84 variable stars have optical counterparts in DSS images. The three diagrams (colour-magnitude, light variance and colour-colour diagrams) indicate that most of the detected variable stars should be largeamplitude and long-period variables such as Mira variables or OH/IR stars. The number density distribution of the detected variable stars implies that they trace the bar structure of the Galactic bulge.