• 제목/요약/키워드: Ordinary Least Squares

검색결과 202건 처리시간 0.025초

A comparison of neural networks to ols regression in process/quality control applications

  • Nam, Kyungdoo;Sanford, Clive C.;Jayakumar, Maliyakal D.
    • 경영과학
    • /
    • 제11권2호
    • /
    • pp.133-146
    • /
    • 1994
  • This study compares the performance of neural networks and ordinary least squares regression with quality-control processes. We examine the applicability of neural networks because they do not require any assumptions regarding either the functional from of the underlying process or the distribution of errors. The coefficient of determination($R^2$), mean absolute deviation(MAD), and the mean squared error(MSE) metrics indicate that neural networks are a viable and can be a superior technique. We also demonstrate that an assessment of the magnitude of the neural notwork input layer cumulative weights can be used to determine the relative importance of predictor variables.

  • PDF

A Linear and Consistent Class of Econometric Estimators in Simultaneous Equations

  • Srivastava, V.K.;Dwivedi, T.D.;Agnihotri, B.S.
    • Journal of the Korean Statistical Society
    • /
    • 제8권2호
    • /
    • pp.117-123
    • /
    • 1979
  • Striaght-forward application of the ordinary least squares model for estimating the parameters of a simultaneous linear stochastic equations model does not provide consistent estimators due to the fact that the explanatory jointly dependent variables are correlated with the disturbances. The search for consistent estimators during the last three decades has yielded a variety of estimators which can be broadly classified into two groups, namely, limited information and full information. Both the groups fails to uilize the over-identifying restrictions in the structural equations except the one under study while the latter group succeeds; see, e.g. Srivastava(1978) for a brief review and Theil (1961) for a detail description.

  • PDF

다변량 시계열 자료의 다중단위근 검정법 (Testion a Multivariate Process for Multiple Unit Roots)

  • Key Il Shin
    • 응용통계연구
    • /
    • 제7권1호
    • /
    • pp.103-112
    • /
    • 1994
  • 본 논문에서는 비정상(단위근) 시계열이 포함된 다변량 시계열 자료에서 단위근에 해당되는 계수행렬 추정량의 극한 분포가 정상시계열의 유무에 상관없이 일정하다는 것을 밝혔다. 또한 단위근만 존재하는 다변량 시계열에서 다중단위근을 검정하는 검정통계량을 제안하였다.

  • PDF

두개의 BLUE가 서로 같을 필요충분조건들과 그 응용 (Necessary and sufficient conditions for the equality between the two best linear unbiased estimators and their applications)

  • 이상호
    • 응용통계연구
    • /
    • 제6권1호
    • /
    • pp.95-103
    • /
    • 1993
  • 두 개의 공분산행렬 $V_1과 V_2$로 구별되는 두 개의 선형모형에서 BLUE끼리 같을 필 요충분조건이 유도된다. 그리고 이 발견으로 쉽게 이해되는 여러 응용사례도 보여준다. 그동 안 여러 논문에서 언급되어 온 BLUE와 OLSE가 같을 필요충분조건도 논의된다.

  • PDF

Large deformation analysis of inflated air-spring shell made of rubber-textile cord composite

  • Tran, Huu Nam;Tran, Ich Thinh
    • Structural Engineering and Mechanics
    • /
    • 제24권1호
    • /
    • pp.31-50
    • /
    • 2006
  • This paper deals with the mechanical behaviour of the thin-walled cylindrical air-spring shell (CAS) made of rubber-textile cord composite (RCC) subjected to different types of loading. An orthotropic hyperelastic constitutive model is presented which can be applied to numerical simulation for the response of biological soft tissue and of the nonlinear anisotropic hyperelastic material of the CAS used in vibroisolation of driver's seat. The parameters of strain energy function of the constitutive model are fitted to the experimental results by the nonlinear least squares method. The deformation of the inflated CAS is calculated by solving the system of five first-order ordinary differential equations with the material constitutive law and proper boundary conditions. Nonlinear hyperelastic constitutive equations of orthotropic composite material are incorporated into the finite strain analysis by finite element method (FEM). The results for the deformation analysis of the inflated CAS made of RCC are given. Numerical results of principal stretches and deformed profiles of the inflated CAS obtained by numerical deformation analysis are compared with experimental ones.

DRGs(Diagnosis Related Groups)별 환자집중도 수준에 따른 입원진료비와 재원일수의 차이 분석 (Is the Hospital Caseload of Diagnosis Related Groups Related to Medical Charges and Length of Stay?)

  • 곽진미;이광수
    • 보건의료산업학회지
    • /
    • 제8권4호
    • /
    • pp.13-24
    • /
    • 2014
  • This study analyzes the effects of hospital caseload on medical charges and length of stay for inpatients. Hospital caseload, representing the level of concentration of patients, was measured with the Internal Herfindal Index for three diagnosis related group (DRG) codes (appendectomy, operations on anus, and operations on uterus and adnexa). Ordinary least squares regression was used for analysis. Results showed that medical charges per inpatient and average length of stay significantly differed with respect to hospital concentration indices, and that hospital caseload was inversely related to operational performance for appendectomy and operations on uterus and adnexa. The significant negative relationship between concentration index and length of stay may decrease the total medical charges. The results imply that the expansion of the DRG payment system to hospitals will have a negative influence on their gross sales.

The Mediating Effects of the Process Design Capability and Product Interior Design Capability on the Relationship between SMEs' External Information Network Diversity and Their New Technology Development Capability

  • Hau, Yong Sauk
    • Asia pacific journal of information systems
    • /
    • 제26권4호
    • /
    • pp.477-488
    • /
    • 2016
  • New technology development capability plays a key role in making small- and medium-sized enterprises (SMEs) increase their innovation performance, such as in product or process innovation. To examine the influencing factors of SMEs' new technology development capability, this study empirically analyzes the mediating effects of SMEs' process design capability and product interior design capability on the positive association between their external information network diversity and new technology development capability. This study performs the ordinary least squares regression on a sample of 2,000 South Korean SMEs. Results reveal that SMEs' process design capability fully mediates, and product interior design capability partially mediates the positive association between the external information network diversity and new technology development capability.

몬테카를로 분석에 의한 효율성 추정방법의 비교 (A Comparison of Efficiency Estimation Methods via Monte Carlo Analysis)

  • 최태성;김성호
    • 경영과학
    • /
    • 제19권1호
    • /
    • pp.117-128
    • /
    • 2002
  • In this Paper we investigate the performance of the five efficiency estimation methods which include the stochastic frontier model estimated by maximum likelihood (SFML), the stochastic frontier model estimated by corrected ordinary least squares (SFCOLS), the data envelopment analysis (DIA) model, the combined estimation of SFML and DEA (SFML + DEA), and the combined estimation of SFCOLS arid DIA (SFCOLS+ DEA) using Monte Carlo analysis. The results include: 1) SFML provides most accurate efficiency estimates for the sample sloe 150 or over,2) SFML+DEAor SFCOLS + DIA Perform better for the cases with sample sloe 25, 50, and low random errors, 3) SFCOLS performs better for the close with sample sloe 25, 50, and very high random errors.

도구변수를 이용한 최저임금의 고용효과 (The Effect of the Minimum Wage on Employment Using Instrumental Variable)

  • 강승복
    • 노동경제논집
    • /
    • 제40권3호
    • /
    • pp.105-131
    • /
    • 2017
  • 이 연구는 정부의 진보성을 도구변수로 이용하여 한국의 최저임금이 고용에 미치는 영향을 분석하였다. 최저임금의 고용효과를 일반적인 최소자승추정법으로 회귀분석하면 변수들 간의 내생성으로 인하여 진정한 효과가 상향 편의될 수 있다. 따라서 적절한 도구변수를 이용하여 내생성을 제거한 인과 관계를 분석할 필요가 있다. 도구변수를 이용한 분석 결과 최저임금 인상률의 증가는 고용을 감소시키는 것으로 나타났다. 이와 같은 최저임금 인상에 따른 음(-)의 고용효과는 신고전학파 경제학에서 예측하는 결과와 부합되는 것이기도 하다.

  • PDF

On the Effect of Estimated Mean for the Weighted Symmetric Estimator

  • Key Il Shin;Hee Jeong Kang
    • Communications for Statistical Applications and Methods
    • /
    • 제4권3호
    • /
    • pp.903-909
    • /
    • 1997
  • The ordinary least squares estimator and the corresponding pivotal statistics have been widely used for the unit test. Recently several test criteria based on maximum likelihood estimators and weighted symmetric estimator have been proposed for testing the unit root hypothesis in the autoregressive processes. Pantula at el. (1994) showed that the weighted symmetric estimator has good power properties. In this article we use an adjusted estimator for mean in the model when we use weighted symmetric estimator. A simulation study shows that for the small samples, this new test criterion has better power properties than the weighted symmetric estimator.

  • PDF