Indeterminate Feature of Parameter Estimation in Multilevel Categorical Latent Variable

Jungkyu Park(박중규) 경북대학교 사회과학기초자료연구소 2019 연구방법논총 Vol.4 No.1

본 논문에서는 다층 잠재계층모형에서 모수의 구조적인 비확정성에 대해서 다룬다. 먼저 다층 잠재계층에서 non-full-rank, independent, permutation indeterminacy의 세 가지 경우에 대해 소개하고 각각의 경우에 발생할 수 있는 모수의 비확정성에 대해서 설명한다. 또한 모수의 비확정성에 대한 이해를 높이기 위해 수리적인 예시를 제시한다. 논의에서는다층 잠재계층모형을 이용해서 자료를 분석할 때 구조적 불확정성에 대한 인식이 중요하다는 점을 강조한다. 덧붙여, 연구자가 선택한 최종모형이 적절한 잠재 구조를 가졌는지를 점검하기 위한 지침을 제공한다. Structural indeterminacy among the multilevel discrete latent variables in the multilevel latent class model (MLCM) is discussed in this paper. Three scenarios - non-full-rank, independent, and permutation indeterminacy - are presented with theoretical explanations and proofs of each structural indeterminate case. Numerical examples are also included to provide intuitive and conceptual understanding of structural indeterminacy. The awareness of the structural indeterminacy in applying the MLCM to data is highlighted in the discussions. Researchers are giving examples and directions to check for problematic structures to ensure their final model has a theoretically sound latent structure when modeling data with multilevel discrete latent variables.

Anodic Performances of Surface-Treated Natural Graphite for Lithium Ion Capacitors

Park, Chul Min,Jo, Yong Nam,Park, Jung Woo,Yu, Ji-Sang,Kim, Jeom-Soo,Choi, Jungkyu,Kim, Young-Jun Korean Chemical Society 2014 Bulletin of the Korean Chemical Society Vol.35 No.9

The surface of natural graphite was modified by the use of hydrogen peroxide and evaluated as an anode material for lithium ion capacitors (LICs). The surface treatment was carried out under various ultrasonic conditions of 200, 300, and 400W, which were applied to a mixture of natural graphite and hydrogen peroxide solution for 1 h. While the bulk structure was maintained, the hexagonal symmetry and physical properties of natural graphite, such as BET surface area, tap density, and particle size, were affected by the surface treatment. FT-IR and XPS measurements confirmed the signature of C=O on the surface of graphite samples after treatment. Both the pristine and surface-treated graphites showed a similar reversible capacity of $370mAhg^{-1}$, and the coulombic efficiency of surface-treated graphite decreased with higher ultrasonic energies (89.1%, 89.0%, and 88.0% for 200, 300, and 400 W) comparing with pristine graphite (89.4%). The capacity retention of LICs was greatly improved with the treated natural graphite. The graphite treated under the ultrasonic energy of 300 W and pristine natural graphite showed capacity retention of 77.5% and 42.9%, implying that the surface treatment was an effective method for the improvement of natural graphite as an anode material for LICs.

Evolution of Luminescence of Sr_{2–<i>y</i>–<i>z</i>}Ca_<i>z</i>Si(O_1–<i>x</i>N_<i>x</i>)₄:<i>y</i>Eu²⁺ with N^3–, Eu²⁺, and Ca²⁺ Sub

Park, Jungkyu,Lee, Seung Jae,Kim, Young Jin American Chemical Society 2013 Crystal Growth & Design Vol.13 No.12

<P>Sr<SUB>2–<I>y</I>–<I>z</I></SUB>Ca<SUB><I>z</I></SUB>Si(O<SUB>1–<I>x</I></SUB>N<SUB><I>x</I></SUB>)<SUB>4</SUB>:<I>y</I>Eu<SUP>2+</SUP> (SCSON:Eu<SUP>2+</SUP>) solid solutions were prepared by substituting N<SUP>3-</SUP>, Eu<SUP>2+</SUP>, and Ca<SUP>2+</SUP> ions into Sr<SUB>2</SUB>SiO<SUB>4</SUB> (SSO). These ions contributed differently to the evolution of luminescence of SCSON:Eu<SUP>2+</SUP>. SSON:Eu<SUP>2+</SUP> (<I>z</I> = 0) has two activation centers: Eu(I) and Eu(II). The nitridation effects led to a dramatic change in the crystal field surrounding the Eu(II) site but rarely affected the Eu(I) site. Accordingly, SSON:Eu<SUP>2+</SUP> exhibited broad excitation spectra from ultraviolet to visible wavelengths. In comparison with the Eu(II) green emissions of SSO:Eu<SUP>2+</SUP>, the dominant peak wavelengths (DPWs) of the Eu(II) emissions were at red emission regions (605–630 nm), depending on the amount of Eu<SUP>2+</SUP> ions. The Ca<SUP>2+</SUP> ions of SCSON:Eu<SUP>2+</SUP> preferentially changed the emission wavelength assigned to Eu(I) and affected the Eu(II) emission intensity. In addition to the spectral properties, the chromaticity coordinates and a low thermal quenching behavior of SCSON:Eu<SUP>2+</SUP> powders demonstrated that they can be a novel red phosphors for use in white light emitting diodes.</P><P>The dramatic red-shift from green to red and the enhanced emission intensity of Sr<SUB>2−<I>z</I></SUB>Ca<SUB><I>z</I></SUB>Si(O<SUB>1−<I>x</I></SUB>N<SUB><I>x</I></SUB>)<SUB>4</SUB>:Eu<SUP>2+</SUP>.</P><P><B>Graphic Abstract</B> <IMG SRC='http://pubs.acs.org/appl/literatum/publisher/achs/journals/content/cgdefu/2013/cgdefu.2013.13.issue-12/cg400751n/production/images/medium/cg-2013-00751n_0011.gif'></P><P><A href='http://pubs.acs.org/doi/suppl/10.1021/cg400751n'>ACS Electronic Supporting Info</A></P>

A comparative study on differential item functioning detection of three approaches to measurement equivalence

Jungkyu Park,Sunho Jung 한국데이터정보과학회 2021 한국데이터정보과학회지 Vol.32 No.1

This article compares three approaches for testing measurement equivalence in terms of detecting differential item functioning. These approaches include mean and covariance structure (MACS), confirmatory factor analysis with ordered categorical variables (ordinal CFA), and item response theory (IRT) model. With the growing interest in measurement invariance in latent variable models, several studies have been mainly conducted to compare the factor analytic approach with the item response theory approach. However, these studies have practical limitations due to unrealistic assumptions about observed variables and parameter nonequivalence. For a meaningful comparison between the factor analytic and item response methods, ordinal CFA is considered another alternative approach to measurement invariance in the present study. The results of the simulation study validate that the ordinal CFA approach presented inflated Type I error rates compared with the MACS and IRT methods in overall conditions, while the statistical power of ordinal CFA were not higher than those of MACS but much higher than those of the IRT model throughout most conditions.

Latent Class Modeling for Nested Data : Introduction to Multilevel Latent Class Model

Jungkyu Park(박중규),Changhwan Shin(신창환) 경북대학교 사회과학기초자료연구소 2020 연구방법논총 Vol.5 No.3

잠재변수를 포함한 통계모형의 기본적인 가정은 개인의 응답이 서로 독립적이라는 것이다. 그러나 관찰치가 상위 수준의 집단에 속한 다층 구조에서 수집된 자료의 경우에는 동일한 집단에 속한 관찰치의 응답 간 상관이 발생하므로 독립성 가정이 충족되지 않는다. 비모수 다층잠재계층 모형은 집단수준의 범주형 잠재변인을 가정하여 동일한 집단에 속하여 발생하는 응답 간 상관을 설명하는 모형이다. 본 논문은 잠재계층분석 모형과 무선효과 모형을 통합한 비모수적 다층잠재계층 모형을 소개하고 실제 자료를 바탕으로 개인과 집단수준에서 최적의 잠재계층의 수를 결정하는 과정에 대해서 논의한다. The fundamental assumption in any latent variable model is that observations are independent of one another, given the latent status. However, this assumption is often inadequate when observations are nested within higher-level units because such nested data structures induce dependencies in data. The nonparametric version of the multilevel latent class model (MLCM) is an extension of latent class model (LCM) in which the dependencies in data are accounted for by discrete latent variables. This paper aims to review models with discrete latent variables and introduce the MLCM which integrates LCM and random effect model. The model selection issue in the MLCM is also discussed with an empirical example.

가솔린 엔진의 헤드와 블록의 열적 및 기계적 거동에 관한 연구

박정규(Jungkyu Park),주원식(Wonseok Joo),김광현(Kwanghyun Kim) 한국자동차공학회 1999 한국 자동차공학회논문집 Vol.7 No.2

Analysis and Modeling on the pH-Dependent Current Drift of Si Nanowire Ion-Sensitive Field Effect Transistor (ISFET)-Based Biosensors

Jang, Jungkyu,Kim, Jungmok,Kim, Seohyeon,Mo, Hyun-Sun,Kim, Dong Myung,Choi, Sung-Jin,Park, Byung-Gook,Kim, Dae Hwan,Park, Jisun American Scientific Publishers 2017 Journal of Nanoscience and Nanotechnology Vol.17 No.5

<P>Although the ion-sensitive field effect transistor (ISFET)-based biosensor has a great potential for point-of-care testing systems, the current drift still remains as a challenging issue for its commercialization. Furthermore, the drift makes the design of readout circuit for a high-resolution biosensor very complicated because it is very sensitive to the amount of ionic species in electrolyte or human serum. However, its chemical/physical origin in neither yet fully understood nor modeled for the circuit design and simulation. In this paper, the mechanism of current drift in ISFET was explained by short term reaction and long term reaction dependent on position of hydrogen ion. Besides, we proposed the analytical drift model of drain current and threshold voltage for different pH levels and various device sizes in the top-down processed SiNW ISFET. We believe that our result is potentially useful for the drift-aware circuit design for high-resolution biosensor system.</P>

시험소간 성능 검증을 위한 벽면 속도 보상 10회선 초음파 유량계 개발

박정규(Jungkyu Park),정진도(Jindo Chung) 한국환경에너지공학회 2023 한국열환경공학회 학술대회지 Vol.2023 No.2

Most wet- type multi-path ultrasonic flowmeters have their measurement accuracy rate to be deteriorated as the flow rate is decreased. Curve-fitting on the characteristics of ultrasonic flowmeters is essential to improve their measurement accuracy rate. This hinders their use as transfer standards in inter-laboratory comparisons for water flow metrology. A breakthrough to enhance the measurement accuracy rate of ultrasonic flowmeters is needed. This study suggests that the concept of near-wall flow velocity should enhance the characteristics of ultrasonic flowmeters. As a result, the repeatability can be improved as low as 土0. 02% which should be suitable to be used as transfer standards. Two main points of using the near-wall flow velocity for ultrasonic flowmeters are emphasized. One is a formula to describe the near-wall flow velocity as a function of both Reynolds number and surface roughness. The other one is the results of comparison calibrations between two national metrology institutes, which can support measurement equivalence. In particular, the number of equivalence is less than one (¦E_n¦ ≤ 1), and the inconclusiveness parameter is less than two (u_comp/u_base≤ 2) if the near- wall flow velocity is applied for ultrasonic flowmeters. This implies that the ultrasonic flowmeters could be used as transfer standards for the forthcoming inter-laboratory comparisons.

방향성 시추기술의 발전 동향 및 적용 사례

박정규(Jungkyu Park),김현태(Hyuntae Kim) 대한기계학회 2014 대한기계학회 춘추학술대회 Vol.2014 No.5-2

고압 완전혼합반응기(PSR)내의 메탄-공기 희박 예혼합 연소의 NO<SUB>x</SUB> 생성에 관한 모델링 연구

박정규(Jungkyu Park),김현(Hyun Kim) 한국자동차공학회 2006 한국자동차공학회 춘 추계 학술대회 논문집 Vol.- No.-

In this study the predictions of NOx in methane-air lean premixed combustion in HP-PSR were carried out by using GRI mech 3.0 and Zeldovich, nitrous oxide, prompt, and NNH NO formation mechanism. Also the comparison of the prediction to HP-JSR experimental data of Rutar for the validation of the model. This study concerns about the importance of the chemical pathways. The chemical pathway most likely to form the NO in lean-premixed combustion was investigated. Results obtained with the 4 different NO mechanisms are compared.