http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
박종성 ( J S Park ),황성준 ( S J Whang ),이승관 ( S K Lee ),이창규 ( C K Lee ),황선철 ( S C Whang ),류정록 ( J R Ryu ),김장영 ( J Y Kim ),강영태 ( Y T Kang ) 대한임상검사과학회 1992 대한임상검사과학회지(KJCLS) Vol.24 No.1
Linear regression analysis is a commonly used tech- in analyzing method comparison data. If a lin- relationship between the test and reference method can be defined, then the slope and intercept of this line can provide estimates of the proportional and constant error between the two methods. ever if the basic assumptions underlying the least squares model are not met, the estimated line may be incorrect. The least-squares regression method may give two very disparate lines, depending on whether X or Y is used as the independant variable. Neither line expresses the functional relationship bet ween the true values of X and Y : both are altered by errors of measurement in the independant variant. In contrast, the slope and the intercept which accurately represented the relation between X and Y could be estimated in linear structural relationship. The Syncron CX5, a random access type analyzer used for routine clinical chemistry determinations, was compared to the Gilford 203S. The analysis of data obtained by running serum samples in parallel on two instruments indicates that the Beckman CX5 does not correlate well with the Gilford 203S.