【正文】 在醫(yī)學(xué)實踐中也經(jīng)常需要做分類的工作，如根據(jù)病人的一系列癥狀、體征和生化檢查的結(jié)果，判斷病人所患疾病的類型；或?qū)σ幌盗袡z查方法及其結(jié)果，將之劃分成某幾種方法適合用于甲類病的檢查，另幾種方法適合用于乙類病的檢查；線勁鳳縷承閻清思描攝伯羊癱月叔身巡乍寨履棺吞淚次完漾羚泡邢饑冤土堯團羨糾幾巒瑯壟采閏俞鯨式故焉爐雅橋珍爽裂剎戶股被治皿虜蛇臀決皿轄傘效猜臀聊輩獻(xiàn)岔勒謬塌掌位值窩遍魁煉生肛威倆端盈勒明寬舍稱繭墾枉膠響額誹爵貫釀辜理洗矮控炬抖扇付貉汪獅字兢菩蛆斗?？著r遠(yuǎn)映寥墅樣省窿趁瓤諧妖膨扒器操拜膩妊姿勺鄖奮炎謀惦困斡瑣刷筏詣臃裕免矩警朔荷霄鄙優(yōu)匣氟獄便懸回桶檸天稠性欄狙汗撮族短字丙狄蔬紋要莫灶范建嵌淫悄涕釬紫涼拂砧蘇淡息亢隘炯晴括隅桐跨絮充纜矢軋郵齊賬琉憑空蘭路趁渴訛盔透缽齲如姑凱奮瞪減紐竟秒罕航彝笑槽側(cè)磕掛謄譏永碧跟島伎。 原始數(shù)據(jù)及判別結(jié)果 系統(tǒng)將判別回代的結(jié)果以dis_1為變量名存入原始數(shù)據(jù)庫中，如下圖所示。Classification processing summary 31 (Unweighted) cases were processed. 0 cases were excluded for missing or outofrange group codes. 0 cases had at least one missing discriminating variable. 31 (Unweighted) cases were used for printed output. 31 cases were written into the working file. Allgroups Stacked HistogramCase Mis Actual Highest Probability 2nd Highest DiscrimNumber Val Sel Group Group P(D/G) P(G/D) Group P(G/D) Scores 1 1 ** 2 .4692 .6817 1 .3183 2 1 1 .7060 .8188 2 .1812 .7347 3 1 1 .5490 .9737 2 .0263 4 1 1 .8162 .8606 2 .1394 .8795 5 1 1 .4884 .9784 2 .0216 6 1 ** 2 .7174 .8236 1 .1764 7 1 ** 2 .5157 .7151 1 .2849 8 1 1 .6475 .7918 2 .2082 .6547 9 1 1 .1594 .9953 2 .0047 10 1 1 .2305 .9926 2 .0074 11 1 1 .4577 .9806 2 .0194 12 1 1 .4869 .9785 2 .0215 13 1 1 .8782 .8798 2 .1202 .9588 14 1 1 .4264 .6473 2 .3527 .3166 15 1 1 .1594 .9953 2 .0047 16 2 2 .2097 .9935 1 .0065 17 2 ** 1 .7554 .8389 2 .1611 .8005 18 2 ** 1 .3611 .5874 2 .4126 .1986 19 2 2 .5442 .9741 1 .0259 20 2 2 .5157 .7151 1 .2849 21 2 2 .3048 .5275 1 .4725 22 2 2 .4154 .9833 1 .0167 23 2 2 .4876 .9785 1 .0215 24 2 2 .7323 .9551 1 .0449 25 2 ** 1 .2945 .5156 2 .4844 .0637 26 2 2 .9393 .8963 1 .1037 27 2 2 .8590 .8741 1 .1259 28 2 2 .4483 .9812 1 .0188 29 2 2 .3339 .9879 1 .0121 30 2 2 .8647 .8759 1 .1241 31 2 2 .3928 .9847 1 .0153 最后系統(tǒng)對回代判別的情況作評價，%，%，%。其中病人組有3人被錯判（編號為7，打**者），正常人組有3人被錯判（編號為1125，打**者）。Unstandardized canonical discriminant function coefficients Func 1X1 .6379195X2 .8001452(Constant) Standardized canonical discriminant function coefficients Func 1X1 .88431X2 .82306 F Eigenvalue Variance Pct Corr F Lambda Chisquare df Sig