• Communications for Statistical Applications and Methods
• /
• 제26권5호
• /
• pp.507-517
• /
• 2019

Statistical matching is a method of combining multiple sources of data that are extracted or surveyed from the same population. It can be used in situation when variables of interest are not jointly observed. It is a low-cost way to expect high-effects in terms of being able to create synthetic data using existing sources. In this paper, we propose the several statistical micro matching methods using a multinomial logistic regression model when all variables of interest are categorical or categorized ones, which is common in sample survey. Under conditional independence assumption (CIA), a mixed statistical matching method, which is useful when auxiliary information is not available, is proposed. We also propose a statistical matching method with auxiliary information that reduces the bias of the conventional matching methods suggested under CIA. Through a simulation study, proposed micro matching methods and conventional ones are compared. Simulation study shows that suggested matching methods outperform the existing ones especially when CIA does not hold.

• 한국수학교육학회지시리즈A:수학교육
• /
• 제58권1호
• /
• pp.139-160
• /
• 2019

This study attempts to examine statistical tasks in the middle-school mathematics textbooks of Korea and Connected Mathematics 3 [CMP] of the US in terms of an opportunity-to-learn for statistical reasoning. We utilized an analytical framework consisting of types of context, statistical reasoning level, cognitive demand of the tasks, and types of student response. The findings from the task analysis suggested that Korean textbooks focused on finding answers by applying previously learned algorithms or formulas and thus provided students with very limited opportunities to experience statistical reasoning. Also, the results proposed that the mathematical tasks in statistics unit of CMP3 offer more essential and complex tasks that promote students' conceptual understanding of various statistical ideas and statistical reasoning in a meaningful way.

• Communications for Statistical Applications and Methods
• /
• 제29권1호
• /
• pp.41-51
• /
• 2022

We forecast the US oil consumption level taking advantage of google trends. The google trends are the search volumes of the specific search terms that people search on google. We focus on whether proper selection of google trend terms leads to an improvement in forecast performance for oil consumption. As the forecast models, we consider the least absolute shrinkage and selection operator (LASSO) regression and the structured regularization method for large vector autoregressive (VAR-L) model of Nicholson et al. (2017), which select automatically the google trend terms and the lags of the predictors. An out-of-sample forecast comparison reveals that reducing the high dimensional google trend data set to a low-dimensional data set by the LASSO and the VAR-L models produces better forecast performance for oil consumption compared to the frequently-used forecast models such as the autoregressive model, the autoregressive distributed lag model and the vector error correction model.

• Journal of the Korean Statistical Society
• /
• 제11권2호
• /
• pp.77-87
• /
• 1982

The choice if constants that define a life testing procedure is considered in terms of the test termination time (censoring time) and the number of items to be tested subject to a given range of variance of the expected life time, where the failure time of life testing is exponentially distributed.

• Journal of the Korean Statistical Society
• /
• 제13권1호
• /
• pp.42-47
• /
• 1984

In this paper, the means of the estimators for the coefficient of variation (CV) in an underlying Pareto distribution are expressed in terms of confluent hypergeometric functions. The numericla values of the biases for the CV estimators in the Pareto distribution are also obtained.

• Communications for Statistical Applications and Methods
• /
• 제5권3호
• /
• pp.869-878
• /
• 1998

In this paper, we propose some jackknife estimators for mean in the exponential model with grouped and censored data. Also, we compare the proposed jackknife estimators to other approximate estimators in terms of the mean square error and bias.

• Communications for Statistical Applications and Methods
• /
• 제5권3호
• /
• pp.879-885
• /
• 1998

In this paper, we propose the approximate maximum likelihood estimators (AMLE) of the location and the scale parameter of the singly left truncated normal distribution. We compare the proposed estimators with the simpler estimators (SE) in terms of the mean squared error (MSE) through Monte Carlo methods.

• Journal of the Korean Statistical Society
• /
• 제26권3호
• /
• pp.383-395
• /
• 1997

This paper proves that the standard bootstrap approximation for the least absolute deviation (LAD) estimate of .beta. in AR(1) processes with infinite variance error terms is asymptotically valid in probability when the bootstrap resample size is much smaller than the original sample size. The theoretical validity results are supported by simulation studies.

• Communications for Statistical Applications and Methods
• /
• 제2권2호
• /
• pp.63-73
• /
• 1995

We consider some robust Bayes estimators using ML-II priors as well as certain empirical Bayes estimators in estimating the finite population mean. The proposed estimators are compared with the sample mean and subjective Bayes estimators in terms of "posterior robustness" and "procedure robustness".re robustness".uot;.

• Journal of the Korean Statistical Society
• /
• 제25권3호
• /
• pp.445-456
• /
• 1996

In this paper, we derive unbiased estimators for the sensitivities of expected performance measures in a Markov renewal process. We restrict our derivation to the performance measures during a busy cycle and apply smoothed perturbation analysis method to find those esti-mators. The results show all the terms in the derived estimators can be obtained from a single sample path.