• Communications for Statistical Applications and Methods
• /
• v.23 no.2
• /
• pp.179-187
• /
• 2016

The fixed effects panel probit model faces "incidental parameters problem" because it has a property that the number of parameters to be estimated will increase with sample size. The maximum likelihood estimation fails to give a consistent estimator of slope parameter. Unlike the panel regression model, it is not feasible to find an orthogonal reparameterization of fixed effects to get a consistent estimator. In this note, a hierarchical Bayesian model is proposed. The model is essentially equivalent to the frequentist's random effects model, but the individual specific effects are estimable with the help of Gibbs sampling. The Bayesian estimator is shown to reduce reduced the small sample bias. The maximum likelihood estimator in the random effects model is also efficient, which contradicts Green (2004)'s conclusion.

• ETRI Journal
• /
• v.31 no.2
• /
• pp.121-128
• /
• 2009

In this paper, we describe a fixed-threshold sequential minimal optimization (FSMO) for structured SVM problems. FSMO is conceptually simple, easy to implement, and faster than the standard support vector machine (SVM) training algorithms for structured SVM problems. Because FSMO uses the fact that the formulation of structured SVM has no bias (that is, the threshold b is fixed at zero), FSMO breaks down the quadratic programming (QP) problems of structured SVM into a series of smallest QP problems, each involving only one variable. By involving only one variable, FSMO is advantageous in that each QP sub-problem does not need subset selection. For the various test sets, FSMO is as accurate as an existing structured SVM implementation (SVM-Struct) but is much faster on large data sets. The training time of FSMO empirically scales between O(n) and O($n^{1.2}$), while SVM-Struct scales between O($n^{1.5}$) and O($n^{1.8}$).

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.46 no.3
• /
• pp.40-47
• /
• 2009

This paper presents new drain bias scheme for the linearity enhancement of envelope tacking power amplifiers for W-CDMA base-stations. In the conventional envelope tracking power amplifiers, the drain bias voltage is lowered close to the knee voltage of transistor, resulting in the severe linearity degradation. To solve this problem, it is proposed in this paper that the amplifier is biased in the conventional class AB mode with a fixed drain bias voltage if the input envelope is low and in the envelope tracking mode otherwise. Moreover, the drain bias in the envelope tracking mode is newly determined to minimized the distortion. To verify the effectiveness of the proposed bias scheme, simulation is performed on the W-CDMA based-station envelope tracking power amplifier using class AB Si-LDMOS power amplifier. It is shown from the simulation that the proposed bias scheme allows a drastic linearity enhancement with the comparable efficiency enough to meet the requirement of W-CDMA base-station without additional linearization techniques.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.9
• /
• pp.41-45
• /
• 2014

In this paper, a high-efficiency power amplifier is implemented by using a drain bias control circuit operated at low input power for WPT(Wireless Power Transfer). Adaptive bias control circuit was added to high-efficiency class-E amplifier. It was possible to obtain the overall improvement in efficiency by adjusting the drain bias at low input power. The proposed adaptive class-E amplifier is implemented by using the input and output matching network and serial resonant circuit for improvement in efficiency. Drain bias control circuit consists of a directional coupler, power detector, and operational amplifier for adjusting the drain bias according to the input power. The measured results show that output powers of 41.83 dBm were obtained at 13.56 MHz. At this frequency, we have obtained the power added efficiency(PAE) of 85.67 %. It was confirmed increase of PAE of an average of 8 % than the fixed bias from the low input power level of 0 dBm ~ 6 dBm.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2007.06a
• /
• pp.162-163
• /
• 2007

The dry etching of Si was investigated using direct dc biasing to the Si substrate. The TCP type etching system with a feed-through for applying a dc bias was used in the etching. The applied dc bias and ICP power was varied to examine the effect on the etching at the fixed chamber pressure and $SF_6$ flow rate of 10 mTorr and 10 sccm during. When the plasma was generated at ICP power of 100 W, the etch rate of Si was increased with the bias for the biased samples. However, the etching of Si for the non-biased sample was enhanced for the increased ICP power.

• Korean Journal of Organic Agriculture
• /
• v.19 no.4
• /
• pp.565-575
• /
• 2011

Recently, the use of antibiotics to improve animal productivity in livestock industry was strictly restricted. For these reason, probiotics have been regarded as one of promising materials for an antibiotic alternative. In this study, we investigated how the probiotics influences on the performance of broiler chicken via meta-analysis. Eighteen researches from 1997 to 2010 were used for meta-analysis. The standard summary effects were calculated via fixed effect model and random effect model (Borenstein et al., 2009). Heterogeneity was calculated by using the Cochran's Q statistics (Kook et al., 2009) and publication bias was calculated via Egger's regression (Lee et al., 2011). In fixed model average daily gain, body weight serum protein content and cecal LAB showed positive effect significantly. Feed intake, feed/gain and serum cholesterol showed significant negative effect. In serum triglyceride, negative effect was found but significance was not shown. In random model, average daily gain body weight and cecal LAB showed positive effects with significance and feed/gain and serum cholesterol represented significant negative effects. Publication bias was found only in feed/gain.

• Journal of Magnetics
• /
• v.5 no.3
• /
• pp.90-98
• /
• 2000

The change in the magnetic properties of a spin-valve multilayer with the structure IrMn (9 m)/CoFe (4 nm)/Cu (2.6 nm)/CoFe (2 nm)/NiEe (6 nm) is investigated as a function of the size and the aspect ratio. At a fixed aspect ratio (the length/width ratio) of 2, the magnetostatic interactions begin to affect the magnetic properties substantially at a spin-valve length of 5 $\mum$, and, at a length of 1 $\mum$, they become even more dominant. In the case of a fixed multilayer size (2.4 $\mum$) which is indicated by the sum of the length and the width, magnetization change occurs by continuous spin-reversal and M-H loops are characterized by no or very small hysteresis at aspect ratios smaller than unity, At aspect ratios greater than unity, magnetization change occurs by spin-flip resulting in squared hysteresis loops. A very large changes in the coercivity and the bias field is observed, and these results are explained by two separate contributions to the total magnetostatic interactions: the coercivity by the self-demagnetizing field and the bias field by the interlayer magnetostatic interaction field.

• The Korean Journal of Applied Statistics
• /
• v.25 no.6
• /
• pp.877-887
• /
• 2012

The Bank of Korea changed its method of GDP estimation from a fixed-weighted to a chain-weighted measure in 2009. The fixed-weighted method had had problems such as substitution bias and the rewriting of economic history. As a result of the change, annual growth rates calculated using the chain-weighted method from 1970 through 2008 turned out to be 0.8%p higher on average than the existing rates. The quarterly average chain-weighted growth rates were 0.19%p higher than the fixed-weighted ones, but they changed in the same directions. In this paper we analyze whether the differences in rates between the two calculation methods would bring about a difference in the cyclical characteristics of GDP. We conclude that although there were differences in growth rates after introduction of the chain-weighted method, there was no difference in the cyclical fluctuation.

• Imaging Science in Dentistry
• /
• v.52 no.2
• /
• pp.133-140
• /
• 2022

Purpose: The aim of this review was to systematically analyze the available literature on the correlation between the gray values (GVs) of cone-beam computed tomography (CBCT) and the Hounsfield units (HUs) of computed tomography (CT) for assessing bone mineral density. Materials and Methods: A literature search was carried out in PubMed, Cochrane Library, Google Scholar, Scopus, and LILACS for studies published through September 2021. In vitro, in vivo, and animal studies that analyzed the correlations GVs of CBCT and HUs of CT were included in this review. The review was prepared according to the PRISMA checklist for systematic reviews, and the risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies tool. A quantitative analysis was performed using a fixed-effects model. Results: The literature search identified a total of 5,955 studies, of which 14 studies were included for the qualitative analysis and 2 studies for the quantitative analysis. A positive correlation was observed between the GVs of CBCT and HUs of CT. Out of the 14 studies, 100% had low risks of bias for the domains of patient selection, index test, and reference standards, while 95% of studies had a low risk of bias for the domain of flow and timing. The fixed-effects meta-analysis performed for Pearson correlation coefficients between CBCT and CT showed a moderate positive correlation (r=0.669; 95% CI, 0.388 to 0.836; P<0.05). Conclusion: The available evidence showed a positive correlation between the GVs of CBCT and HUs of CT.

• Communications for Statistical Applications and Methods
• /
• v.19 no.1
• /
• pp.23-32
• /
• 2012

This paper deals with the bias and variance of regression coefficient estimators in a finite population. We derive approximate formulas for the bias, variance and mean square error of two estimators when we select a fixed-size inclusion probability proportional to the size sample and then estimate regression coefficients by the ordinary least square estimator as well as the weighted least square estimator based on the selected sample data. Necessary and sufficient conditions for the comparison of the two estimators in terms of variance and mean square error are suggested. In addition, a simple example is introduced to numerically compare the variance and mean square error of the two estimators.