Method Of Statistical Inference For Interrelative Sample

Because of the intensive feeding and locked breeding system, there usually withcertain relationShip lies in exPerimental individuals in animal production and breeding.If it has certain heritability of experimental index or trait, with resulting, there will becertain interrelation betWeen exPerimental numeral data or phynotyPic values. ThesamPle Which with interrelation betwen random vedables is called interrelativesamPle. And under this definihon, following item were stUdied in thes paPer.l. By the theory of quantitave genetics, divided the random error into tWocomPonens, the diviation of breeding value in experimental individual and theenvironmental error. Then caculated the variance of random error and establisheda mathematical model to proceed statistical inference for interrelative samPle.2. Studied the genera method of the staistica1 inference for interrelative samPleconditional on the known and unknown heritability of experimental index or traitusing general least square and REML.3. According to the practical animal experiment, deduced the simPlized methodcorresPonding to twO designs, one-way randomized classification or block design,for there being particular relationship betWen individuals.4' Regarding block (litter) as random effect, estimated the heritability of experimentalindex by Henderson method, and put forward the method of statistical inferencefOr the interrelative sample from the data of one-way randomized block design.5. ProgranUned a routine for general method.The method above can be used in animal experAnent and nutrition experiment. Ifthe relationship between animals is known, then the statistic information of two parts, nutrition and genetics, can be obtained by analyzing the variation of experimental index. And the crossing and infusion of animal nutrition and genetics will be possible.