• Communications for Statistical Applications and Methods
• /
• v.23 no.6
• /
• pp.575-585
• /
• 2016

Longitudinal studies repeatedly measure outcomes over time. Therefore, repeated measurements are serially correlated from same subject (within-subject variation) and there is also variation between subjects (between-subject variation). The serial correlation and the between-subject variation must be taken into account to make proper inference on covariate effects (Diggle et al., 2002). However, estimation of the covariance matrix is challenging because of many parameters and positive definiteness of the matrix. To overcome these limitations, we propose autoregressive moving average Cholesky decomposition (ARMACD) for the linear mixed models. The ARMACD allows a class of flexible, nonstationary, and heteroscedastic models that exploits the structure allowed by combining the AR and MA modeling of the random effects covariance matrix. We analyze a real dataset to illustrate our proposed methods.

• Journal of Korea Technical Association of The Pulp and Paper Industry
• /
• v.34 no.2
• /
• pp.73-83
• /
• 2002

Gloss is very important optical property influencing the perceived quality of the paper surface as well as the surface after printing. Although the average gloss level of paper products or printed images is important to meet end use specifications, the occurrence of gloss mottle, or non-uniformity of gloss, is often of greater concern for meeting quality requirements, especially for the high gloss paper. Gloss variation originates from the irregularities of paper surface, especially surface roughness of paper. Roughness of paper can be divided into micro-roughness (under $1\mu m$ scale in variation) and macro-roughness (over $1\mu m$ scale in variation) depending on the scale of the irregularities. A clearer understanding of the gloss variation of paper can be achieved by separating the contributions of these two scales of roughness, and characterizing them independently. In order to do this, a novel gloss measuring method was introduced. This can detect local gloss with very high resolution. The effect of macro-roughness on gloss variation, which was identified by the measurable surface topography, was separated from the total gloss variation by using this method. The effect of micro-roughness was then estimated indirectly. The local gloss variations of various paper samples were then evaluated to demonstrate the utility of this approach.

• Korean Journal of Plant Resources
• /
• v.29 no.6
• /
• pp.679-689
• /
• 2016

Lettuce is an important annual leafy vegetable and bitterness is its potent flavor character. Lettuce germplasm differ their phenotypic characters and sesquiterpene lactones (SLs) contents which are important for consumer's acceptance. This study was carried out to evaluate the phenotypic characters and SLs contents in one hundred lettuce germplasm in Jeonju, Korea. Twenty-three agro-morphological (16 qualitative and 7 quantitative) traits and two SLs (lactucin and lactucopicrin) contents were studied in these germplasm. Germplasm exhibited the variation in qualitative and quantitative characters. Average plant weight was 423.9 g with a range from 116.0 to 905.0 g. Lactucin content was varied from 19.7 (IT 294226) to $194.4{\mu}g/g$ (IT 294298) with an average concentration of $84.7{\mu}g/g$. Lactucopicrin ranged from 82.5 (IT 300134) to $2228.6{\mu}g/g$ (IT 294210) with an average concentration of $586.3{\mu}g/g$. Total SLs content was ranged from 120.1 (IT 300134) to 2286.6 (IT 294210)${\mu}g/g$ with the average concentration of $671.0{\mu}g/g$. Significant ($p{\leq}0.05$) differences were found between crisp head and butter head germplasm for lactucin, lactucopicrin and total SLs content. Crisp head germplasm revealed the highest average lactucin ($112.9{\mu}g/g$), lactucopicrin ($734.8{\mu}g/g$) and total SLs content ($847.7{\mu}g/g$). Crisp head and leafy type germplasm exhibited more total SLs content (847.7 and $744.7{\mu}g/g$, respectively) than cos ($524.9{\mu}g/g$) and butter head type ($519.4{\mu}g/g$). Principal component analyses of the quantitative traits indicated that the first principal component axis accounted more than 91% of the total variation. This study revealed the ample genetic variation in the agro-morphological traits and SLs contents to support the selection for improved lettuce varieties.

• Korean Journal of Veterinary Service
• /
• v.17 no.3
• /
• pp.208-226
• /
• 1994

From January to December 1993, 3,385 bulk milk samples were collected from 293 herds in Inchon area. Standard plate counts(SPC) and somatic cell counts(SCC) were performed by the method of milk collection, the situation of raising management and monthly. The results obtained are summarized as follows : 1, Annual average SPC and SCC were respectively 638,000 cfu /mm and 647,000 cells /mm. SPC showed an abrupt decrease from January-1,088,000 cfu /mm to December-279,000 cfu /mm, but SCC showed a slow change from January-1,017,000 cells /mm to December -673,000 cells /mm 2. Variation on milk quality(annual average SPC) was shown a wide difference between everyday collection-575,000 cfu /mm and every other day collection-1,243,000 cfu /mm ac-cording to frequency of milk collection from dairy farms. However, there was a little difference In SCC. 3. In the raising scale, average SPC were the lowest in 16~25mi1king cows, and average SCC were the lowest in above 25milking cows. 4. According to types of milking machine, average SPC and SCC of dairy farms that are equipped with pipeline system were respectively 361,000 cfu /mm and 591,000 cells /mm. Those of dairy farms with bucket system were 549,000 cfu /me and 559,000 cells /mm. 5. In the types of management, average SPC an SCC of dairy farms with hired herdsman were 288,000 cfu /mm and 559,000 cells /mm. Those of dairy farms with self-management were 526,000 cfu /mm and 568,000 cells /mm.

• Journal of the Korean Society of Tobacco Science
• /
• v.26 no.2 s.52
• /
• pp.85-92
• /
• 2004

This study was conducted to investigate the relationship between weather factors during the growing season and chemical components of burley tobacco. Chemical components used in this study was from 'Farm Leaf Tobacco Test' conducted at KT&G Central Research Institute from 1987 through 2002. Data of weather factors during growing season(April to July) were collected in 6 districts measured at Korea Meteorological Adminstration(KMA). Total nitrogen content was increased from 1987 through 2002. Year to year variation of rainfall was the largest, followed by that of sunshine hour. Month to month variation of rainfall also was the largest, followed by that of mean daily air temperature. A negative correlation was found between rainfall and sunshine hour. Relative humidity(R.H.) was correlated positively with rainfall, whereas negatively with sunshine hour. The negative correlations were found between nicotine content and rainfall(in June, May$\~$June, June$\~$July, May$\~$July and average), and R.H.(in June, May$\~$June, June$\~$July, May$\~$July and average), respectively. The negative correlations were found between crude ash content and rainfall(in June and May$\~$June), and R.H.(in June, May$\~$June, June$\~$July and May$\~$July), respectively. Ether extraction content was correlated positively with mean daily air temperature(in July, June $\~$July and May$\~$July) and with sunshine hour(in July, June$\~$July and May$\~$July), but negatively with rainfall(average) and with R.H.(in April, July, June$\~$July, April$\~$June, May­July and average), respectively. Chloride content was correlated positively with sunshine hour(in May, April$\~$May, May$\~$June, April$\~$June, May$\~$July and average), but negatively with rainfall(in June, May$\~$June, June$\~$July, April$\~$June, May$\~$July and average).

• International Union of Geodesy and Geophysics Korean Journal of Geophysical Research
• /
• v.22 no.1
• /
• pp.24-32
• /
• 1994

A special upper-air observation including airsonde and pibal observations was performed to investigate the characteristics features of the vertical distribution of the meteorological elements over Taegu on a selected clear day of each season from October 1991 to August 1992. The diurnal and seasonal variations of the vertical profiles of air temperature and mixing ratio were obtained from airsonde observations and wind speed and direction from pibal observations. The results of these special upper-air observations are as follow : The diurnal variation of the vertical distribution of air temperature reveals the characteristic features associated with the atmospheric boundary layer. All case days, except for the summer season, show upper-level inversion layer which influenced by surface high, and surface inversion layer produced by radiative cooling. The diurnal variation of mixing ratio shows the maximum vale at 1500 LST in both the upper and low levels, and is larger on the lower level than the upper level. The mixing ratio of the lower level is larger than that of the upper level. On the average the mixing ratio decrease with the height, and is the wettest on the summer case day and the driest on the winter case day. The diurnal variation of the wind velocity and direction are variable in the lower level with time and height, while they are steady in the upper level. On the average, the wind direction is southerly or southeasterly for the summer case day, westerly or northwesterly for the spring and fall case days, and northerly or northwesterly for the winter case day.

• Proceedings of the Korean Environmental Health Society Conference
• /
• 2001.11a
• /
• pp.6-10
• /
• 2001

Distribution characteristics, variation patterns and affecting factors of hetorotrophic bacteria were studied from April to May 1999 in Bohai Sea by standard Acridine Orange epifluorescence microscopy (AO method). The biomass in surface waters showed a small day-night variation, varying from 0.13-2.51$\mu\textrm{m}$$.$dm$\^$-3/ with an average of 0.84 $\mu\textrm{m}$$.$dm$\^$-3/. The biomass in bottom waters showed, however, a large variation, changing from 0.15-4.18 $\mu\textrm{m}$$.$dm$\^$-3/ with an average of 1.36 $\mu\textrm{m}$$.$dm$\^$-3/. The peak values were obtained at 5 and 11 am. The bottom water biomass showed a significant correlation with particulate organic carbon (r=0.639, p<0.05). Heterotrophic bacteria showed high biomass in nearshore waters and low values in offshore areas with a high biomass zone around Yellow Sea river mouth, which was consistent with the distribution of nutrients. The vertical distribution of heterotrophic bacteria showed biomass in bottom waters was higher than in surface water. The biomass of heterotrophic bacteria in Bohai Sea was similar with that in other marine waters.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.2098-2114
• /
• 2021

Light intensity variation is one of the key factors which affect the accuracy of vehicle face re-identification, so in order to improve the robustness of vehicle face features to light intensity variation, a Nonnegative Matrix Factorization model with the constraint of image acquisition time difference is proposed. First, the original features vectors of all pairs of positive samples which are used for training are placed in two original feature matrices respectively, where the same columns of the two matrices represent the same vehicle; Then, the new features obtained after decomposition are divided into stable and variable features proportionally, where the constraints of intra-class similarity and inter-class difference are imposed on the stable feature, and the constraint of image acquisition time difference is imposed on the variable feature; At last, vehicle face matching is achieved through calculating the cosine distance of stable features. Experimental results show that the average False Reject Rate and the average False Accept Rate of the proposed algorithm can be reduced to 0.14 and 0.11 respectively on five different datasets, and even sometimes under the large difference of light intensities, the vehicle face image can be still recognized accurately, which verifies that the extracted features have good robustness to light variation.

• Journal of the Korean Home Economics Association
• /
• v.30 no.4
• /
• pp.15-26
• /
• 1992

To investigate the seasonal effects on physiological responses of human body, clothing micro-climate, and subjective sensation, selected the cloths the most frequently dressed by men in spring and fall, and completed wearing trials in the climatic chamber. The results are as follows: 1. Rectal temp. ranged 36.8-37.1$^{\circ}C$ in either spring or fall, and no seasonal effect was found. 2. In skin temp., there was no seasonal effect in forehead, abdomen, and forearm. Skin temp. of chest was higher in spring than in fall. On the contrary, reverse was true in high and leg. Average skin temp. ranged 32.2-33.2$^{\circ}C$ in spring and 32.9-34.$0^{\circ}C$ in fall. 3. Average total sweat rate of spring, 79.4g/hr, was smaller than that of fall, 110.9g/hr. 4. Clothing temp. ranged 28.1-32.8$^{\circ}C$ in spring and 27.6-31.$0^{\circ}C$ in fall. Clothing humidity ranged 36.9-48.9% in spring and 38.2-51.1% in fall. Therefore, clothing microclimate was higher during fall than during spring. As results, skin temp. of the body core except chest did not show seasonal variation, but there was obvious seasonal variation in skin temp. of the extremities. Therefore, seasonal variation should be take into consideration in the experiments related to the cloth. In addition, standard for each season and the degree of work performance should be re-established in clothing micro-climate.

• Journal of the Korean Solar Energy Society
• /
• v.28 no.3
• /
• pp.45-52
• /
• 2008

In order to clarify characteristics of variation in wind energy over a long-term period, an investigation was carried out at Seoguang site on Jeju island. The wind data for 10 years from Automatic Weather System (AWS) were analyzed for each year. The variation in the annual energy production (AEP) for the 2 MW wind turbine was estimated through statistical work. The result shows that the range of the yearly average wind speed at 15 m above ground level for 10 years was from -22.6% to +13.7%, which is wider range than that in Japan. The coefficient of variation for the AEP was 22.7%, which is about twice of that for the yearly average wind speed. Therefore, for estimating the wind energy potential accurately at a given site, the wind data should be analyzed over a long-term period based on the data from the meteorological station.