• Proceedings of the Korean Statistical Society Conference
• /
• 2003.10a
• /
• pp.253-256
• /
• 2003

In ecological studies, animal science, or entomology, the variance of count is considered to have the power of the mean relationship with the mean count as Taylor (1961) presented his famous 'Taylor's Power Law'. In this talk, we are going to review the development of TPL and its extension toward pest management sampling scheme. Different estimation methods are compared. Quasilikelihood approach is suggested to incorporate covariate information. Possible extensions will be discussed.

• Journal of the Korean Geotechnical Society
• /
• v.33 no.7
• /
• pp.29-41
• /
• 2017

There are uncertainties about the seismic load caused by seismic waves, which cannot be predicted due to the characteristics of the earthquake occurrence. Therefore, it is necessary to consider these uncertainties by probabilistic analysis. In this paper, procedures to develop a fragility curve that is a representative method to evaluate the safety of a structure by stochastic analysis were proposed for cut slopes. Fragility curve that considers uncertainties of soil shear strength parameters was prepared by Monte Carlo Simulation using pseudo static analysis. The fragility curve considering the uncertainty of the input ground motion was developed by performing time-history seismic analysis using selected 30 real ground input motions and the Newmark type displacement evaluation analysis. Fragility curves are represented as the cumulative probability distribution function with lognormal distribution by using the maximum likelihood estimation method.

• Journal of the Korean Data and Information Science Society
• /
• v.9 no.2
• /
• pp.95-103
• /
• 1998

Azzalini and Bowman proposed the pseudo-likelihood ratio test for checking the linear relationship using kernel regression estimator when the error of the regression model follows the normal distribution. We modify their method with the bootstrap technique to construct a new test, and examine the power of our test through simulation. Our method can be applied to the case where the distribution of the error is not normal.

• The Korean Journal of Applied Statistics
• /
• v.8 no.2
• /
• pp.133-145
• /
• 1995

This paper is concerned with the method of estimating lifetime distribution from field data for repairable products with multiple modes of failure, and is an extension of Bai et al.(1995). The log linear function is considered as a model for describing the relation between failure time of a product and covariates. Using the nonhomogeneous poisson process, general methods for obtaining pseudo maximum likelihood estimators(PMLEs) for the parameters are outlined and specific formulas for Weibull distribution are obtained. Effects of follow-up percentage on the PMLEs are investigated. Extension to case-cohort design is also considered.

• Communications for Statistical Applications and Methods
• /
• v.15 no.6
• /
• pp.993-1002
• /
• 2008

Correlated responses have been widely analyzed since Liang and Zeger (1986) introduced the famous Generalized Estimating Equations(GEE). However, their variance functions were restricted to known quantifies multiplied by scale parameter. In so many industries and academic/research fields, power-of-the-mean variance function is one of the common variance function. We suggest GEE-type pseudolikelihood estimation based on the power-of-the-mean variance using existing software and investigate it's efficiency for different working correlation matrices.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.6
• /
• pp.1852-1860
• /
• 2000

Distinct from the Markov random field or pseudo 2D HMM models for image analysis, this paper proposes a new model of 2D hidden Markov mesh chain(HMMM) model which subsumes the definitions of and the assumptions underlying the conventional HMM. The proposed model is a new theoretical realization of 2D HMM with the causality of top-down and left-right progression and the complete lattice constraint. These two conditions enable an efficient mesh decoding for model estimation and a recursive maximum likelihood estimation of model parameters. Those algorithms are developed in theoretical perspective and, in particular, the training algorithm, it is proved, attains the optimal set of parameters.

• Journal of Applied Reliability
• /
• v.2 no.2
• /
• pp.63-83
• /
• 2002

Degradation data can provide more reliability information than traditional failure-time data, especially products with few or no failures. This paper is concerned with a method of estimating lifetime distribution from field data with supplementary information on degradation data and covariates. When a distribution of degradation rate obtained by follow-up study for a portion of products that survive after-warranty follows a reciprocal-Weibull or lognormal distribution. A time-to-failure distribution of the product follows Weibull or lognormal distribution, respectively. A method of estimating lifetime parameters for this kind of data and their asymptotic properties are studied. Effects of after-warranty report probability, follow-up rate, and proportion of degradation data on pseudo maximum likelihood estimators of these parameters are investigated.

• East Asian Economic Review
• /
• v.18 no.1
• /
• pp.83-107
• /
• 2014

This paper studies the effects of economic development and cultural proximity as common determinants of trade in cultural goods in a dynamic preference selection model. For the empirical analysis, this paper utilizes the gravity framework with importer fixed effects and Poisson pseudo-maximum likelihood estimators. This paper applies the model to Korean export of broadcasting contents to Asian countries. The relative economic development of the export country and the market size of the import country are important determinants of cultural trade, the results of which are generally consistent with traditional goods trade. However, the distance variable does not show much significance, reflecting the unique characteristics of trade in cultural goods.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.3
• /
• pp.16-23
• /
• 2001

In this paper, we propose the pseudo n-gram language models for speech recognition with middle size vocabulary compared to large vocabulary speech recognition using the statistical n-gram language models. The proposed method is that it is very simple method, which has the standard structure of ARPA and set the word probability arbitrary. The first, the 1-gram sets the word occurrence probability 1 (log likelihood is 0.0). The second, the 2-gram also sets the word occurrence probability 1, which can only connect the word start symbol and WORD, WORD and the word end symbol . Finally, the 3-gram also sets the ward occurrence probability 1, which can only connect the word start symbol , WORD and the word end symbol . To verify the effectiveness of the proposed method, the word recognition experiments are carried out. The preliminary experimental results (off-line) show that the word accuracy has average 97.7% for 452 words uttered by 3 male speakers. The on-line word recognition results show that the word accuracy has average 92.5% for 20 words uttered by 20 male speakers about stock name of 1,500 words. Through experiments, we have verified the effectiveness of the pseudo n-gram language modes for speech recognition.

• The Korean Journal of Applied Statistics
• /
• v.8 no.1
• /
• pp.89-104
• /
• 1995

This paper is concerned with the method of estimating lifetime distributin from field data for products with multiple modes of failure. When product failures occur within warranty period, a manufacturer can obtain failure-record data; failure times, causes of failure, and covariates. Since these data are seriously incomplete for satisfactory inference, that is, only failures occured during warrantly period may be recorded, it is usually necessary to incoporate the failure-record data by taking a supplementary sample of items obtained following up a portion of products that survive warranty time. The log linear function is considered as a model for describing the relation between failure time of a product and covariates. General methods for obtaining pseudo maximum likelihood estimators(PMLEs) for the parameters are outlined and their asymptotic properties are studied, and specific formulas for exponential or Weibull distribution are obtained. Effects of follow-up percentage on the PMLEs are investigated. Extensions to calendar time warranty or calendar and obtaining time warranty are also considered.