• Journal of the Korean Statistical Society
• /
• v.35 no.4
• /
• pp.453-458
• /
• 2006

It is well known fact for the iid data that the limiting standard errors of sample autocorrelations are all unity for all time lags and they are asymptotically independent for different lags (Brockwell and Davis, 1991). It is also usual practice in time series modeling that this fact continues to be valid for white noise series which is a sequence of uncorrelated random variables. This paper contradicts this usual practice for white noise. We consider a sequence of martingale differences which belongs to white noise time series and derive exact joint asymptotic distributions of sample autocorrelations. Some implications of the result are illustrated for conditionally heteroscedastic time series.

• Journal of the Korean Statistical Society
• /
• v.17 no.1
• /
• pp.24-29
• /
• 1988

The asymptotic distribution of the sample partial autocorrelation terms after lag p of an AR(p) model has been shown by Dixon(1944). The derivation is based on multivariate analysis and looks tedious. In this paper we present an interesting contour-integral derivation.

• The Korean Journal of Applied Statistics
• /
• v.20 no.3
• /
• pp.499-514
• /
• 2007

It is a common practice to use the auxiliary variables to impute missing values from item nonresponse in surveys. Sometimes there are few auxiliary variables for missing value imputation, but if spatial and time autocorrelations exist, we should use these correlations for better results. Recently, Lee et al. (2006) showed that spatial autocorrelation could be efficiently used for missing value imputation when spatial autocorrelation existed, using the data from the farm household economy data in Gangwon-do, 2002. In this paper, we present au evaluation of spatial and space-time nonresponse imputation methods when there exist spatial and time autocorrelations using the monthly data during 2000-2002 from the same data previously used by Lee et al. (2006). We show that space-time imputation method is more efficient than the other through the numerical simulations.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.2
• /
• pp.559-564
• /
• 2016

The Durbin-Watson (DW) test in regression model and the Ljung-Box (LB) test in ARMA (autoregressive moving average) model are typical examples of correlated error tests. The DW test is used for detecting autocorrelation of errors using the residuals from a regression analysis. The LB test is used for specifying the correct ARMA model using the first some sample autocorrelations based on the residuals of a tted ARMA model. In this article, simulations with four data generating processes have been carried out to evaluate their performances as correlated error tests. Our simulations show that the DW test is severely dependent on the assumed AR(1) model but isn't sensitive enough to reject the misspecified model and that the LB test reports lackluster performance in general.

• Management & Information Systems Review
• /
• v.30 no.2
• /
• pp.257-275
• /
• 2011

The price relationship between the futures market and the underlying spot market has attracted the attention of academics, practitioners, and regulators due to their roles during periods of turbulence in financial markets. The purpose of this paper is to investigate the dynamic of price relationship(or lead-lag relationship) between Korean Treasury Bond futures market and spot market. To examine the nature of the price relationship, descriptive statistics, serial correlation, and cross-correlation are used as a preliminary statistics in the Korean Treasury Bond spot and futures market. Next, following Stoll-Whaley(1990) and Chan(1992), the multiple regression method is used to examine the lead-lag patterns between the two markets. The empirical results are summarized as follows. The mean returns of spot markets and future markets are positive(+) and negative(-) respectively and the standard deviation of both stock and futures returns increase through the sub-periods. For the most periods, there is negative skewness in the both markets. The zero excess kurtosis due to the heavy tails of the distribution are relatively large. The autocorrelations in the spot returns for the sample periods are positive in time lag 1, but the autocorrelations in the future returns shows no significant evidence. The results of the daily cross-correlations between the KTB spot and futures returns indicate that a lead-lag relationship don't exist for price changes of futures and spot markets as a preliminary analysis. Finally, empirical results of regression analysis for both market indicate that there is no evidence that the KTB futures lead the KTB spot market, or the KTB spot market lead the KTB futures market. These results are robust for all sub-periods.

• Journal of Korean Society of Forest Science
• /
• v.96 no.6
• /
• pp.667-675
• /
• 2007

The genetic diversity and the spatial structure in two populations of Abelia tyaihyoni in Yeongwol region were studied by employing I-SSR markers. In spite of the limited distribution and small population sizes of Abelia tyaihyoni, the amount of genetic diversity estimated at the individual level was comparable to other shrub species (S.I.=0.336, h=0.217). Genetic diversity at the genet level was very similar to that at individual level. (S.l.=0.339, h=0.219). About 18.7 percent of total variation was allocated between two populations, which was slightly higher or similar level as compared with other shrub species. Genotypic diversity estimated by the ratio of the number of genets ($N_G$) over the total number of individuals (N) and a modified Simpson's index ($D_G$) were also higher than those of other shrubs. The maximum diameter of a genet did not exceed 5.5 m. The high level of gene and genotypic diversity, and the relatively limited maximum diameter of a genet suggested that the clonal propagation is not the most dominant factor in determining the population structure of Abelia tyaihyoni. Spatial autocorrelation analysis revealed significant spatial genetic structure within 12 m and 18 m distances in two populations A and B, respectively. Autocorrelations among individuals at the both individual and genet levels in each population didn't show any considerable differences. As a sampling strategy for ex-situ conservation of populations showing continuous distribution, a minimum distance of 18 m between individuals was recommended. For the populations with many segments, it was considered very crucial to sample materials from as many segments as possible.