META BINARY
META BINARY
は、Statistics Base Edition で使用できます。 シンタックスは、効果サイズの推定のためにアクティブなデータ・セットに生データが提供されている場合の、2 値結果のメタ分析プロシージャーを表します。
META BINARY
[ /CRITERIA
[ SCOPE = {AVAILABLE**} {LISTWISE} ]
[ CLASSMISSING = {EXCLUDE**} {INCLUDE} ]
[ CILEVEL = {95**} {value} ]
[ MAXITER = {100**} {integer} ]
[ MAXSTEP = {5**} {integer} ]
[ CONVERGENCE = {1E−6**} {value} ]
[ DROP = {FALSE**} {TRUE}]
[ ZEROCOUNTS = {ONLY**} {IFALL} {ALL} {KEEP} ]
[ ADD = {0.5**} ]
]
/DATA
TREATMENT = SUCCESS({variable}) FAILURE({variable})
CONTROL = SUCCESS({variable}) FAILURE({variable})
[ ESTYPE = {LOGOR**} {LOGPETO} {LOGRR} {RD} ]
[ ID = {variable} ]
[ STUDY = {variable} ]
[ /ANALYSIS
[ SUBGROUP = {variable} ]
[ CUMULATIVE = {variable}[ (SORT ={ASCENDING**} {DESCENDING}) ] ]
]
[ /INFERENCE
[ MODEL = {RANDOM**} {FIXED[({INV_VAR**}{MANTEL_HAENSZEL})]} ]
[ ESTIMATE = {REML**} {ML} {BAYES}
{HEDGES} {HUNTER_SCHMIDT} {DERSIMONIAN_LAIRD} {SIDIK_JONKMAN}
]
[ ADJUSTSE = {NONE**} {KNAPP_HARTUNG} {TRUNCATED_KNAPP_HARTUNG} ]
]
[ /CONTRAST
[ VARIABLES = {variable_list} ]
[ COEFFICIENTS = {values} ]
[ EFORM = {FALSE**} {TRUE} ]
]
[ /BIAS
TEST = [EGGER] [HARBORD] [PETERS]
[ COVARIATES = {covariate_list} ]
[ FACTORS = factor_list ]
[ INTERCEPT = {INCLUDE**} {EXCLUDE} ]
[ MULTIPLICATIVE = {FALSE**} {TRUE} ]
[ DISTRIBUTION = {T**} {NORMAL}]
]
[ /TRIMFILL
[ SIDE = {EGGER_SLOPE**} {LEFT} {RIGHT} ]
[ METHOD = {LINEAR**} {RUN} {QUADRATIC} ]
[ MODEL = {RANDOM**} {FIXED} ]
[ ESTIMATE = {REML**} {ML} {BAYES}
{HEDGES} {HUNTER_SCHMIDT} {DERSIMONIAN_LAIRD} {SIDIK_JONKMAN} ]
[ ADJUSTSE = {NONE**} {KNAPP_HARTUNG} {TRUNCATED_KNAPP_HARTUNG} ]
]
[ /PRINT
[ HOMOGENEITY] [ HETEROGENEITY ] [ INDIVIDUAL ] [ CUMULATIVE ] [ PREDICTION ]
[ EFORM({FALSE**} {TRUE}) ]
]
[ /SAVE
[ ES[(var_name)] [ ES_EXP[(var_name)] ] [ SE_ES[(var_name)] ]
[ CIL_ES[(var_name)] [ CIL_ES_EXP[(var_name)] ]
[ CIU_ES[(var_name)] [ CIU_ES_EXP[(var_name)] ]
[ PVAL_ES[(var_name)] ] [ WEIGHT[(var_name)] ] [ WEIGHT_PCT[(var_name)] ]
]
[ /OUTFILE
CUMSTATS({'savfile'} {dataset})
[ CUMES[(var_name)] ] [ CUMES_EXP[(var_name)] ] [ SE_CUMES[(var_name)] ]\
[ CIL_CUMES[(var_name)] ] [ CIL_CUMES_EXP[(var_name)] ]
[ CIU_CUMES[(var_name)] ] [ CIU_CUMES_EXP[(var_name)] ]
[ PVAL_CUMES[(var_name)] ]
]
[ /FORESTPLOT
[ DISPLAY = {[ES] [SE] [CI] [WEIGHT] [PVAL]} ]
[ EFORM = {FALSE**} {TRUE} ]
[ ADDCOLS = {variable_list} ]
[ POSITION = {RIGHT**} {LEFT} ]
[ SORT = {variable}[({ASCENDING**} {DESCENDING})] ]
[ REFLINES = {[OVERALL] [NULL]} ]
[ ANNOTATIONS = {[HOMOGENEITY] [HETEROGENEITY] [TEST]} ]
[ CROP = {value1 value2} ]
]
[ /CUMFORESTPLOT
[ DISPLAY = {[ES] [SE] [CI] [PVAL]} ]
[ EFORM = {FALSE**} {TRUE} ]
[ ADDCOLS = {variable_list} ]
[ POSITION = {RIGHT**} {LEFT} ]
[ CROP = {value1 value2} ]
]
[ /BUBBLEPLOT
PREDICTORS = {variable_list}
[ CENTER = {FALSE**} {TRUE} ]
[ PROPORTION = {TRUE**} {FALSE} ]
[ FITLINE = {TRUE**} {FALSE} ]
[ CI = {TRUE**} {FALSE} ]
[ LABEL = {variable}[({AUTO**} {RIGHT} {LEFT} {UPPER} {BOTTOM})] ]
[ YRANGE = {value1 value2} ]
[ XRANGE = {value1 value2} ]
]
[ /FUNNELPLOT
[ IMPUTE = {FALSE**} {TRUE[(REFLINE)]} ]
[ YAXIS = {SE**} {INV_SE} {VAR} {INV_VAR} ]
[ LABEL = {variable}[({AUTO**} {RIGHT} {LEFT} {UPPER} {BOTTOM})] ]
[ YRANGE = {value1 value2} ]
[ XRANGE = {value1 value2} ]
]
[ /GALBRAITHPLOT
[ CI = {TRUE**} {FALSE} ]
[ LABEL = {variable}[({AUTO**} {RIGHT} {LEFT} {UPPER} {BOTTOM})] ]
[ YRANGE = {value1 value2} ]
[ XRANGE = {value1 value2} ]
]
[ /LABBEPLOT
[ PROPORTION = {TRUE**} {FALSE} ]
[ NULL = {TRUE**} {FALSE} ]
[ OVERALL = {FALSE**} {TRUE} ]
[ LABEL = {variable}[({AUTO**} {RIGHT} {LEFT} {UPPER} {BOTTOM})] ]
[ YRANGE = {value1 value2} ]
[ XRANGE = {value1 value2} ]
]
* * サブコマンドが省略された場合のデフォルト。
このコマンドは、アクティブ・データ・セットを読み取り、保留中のコマンドを実行させます。 詳しくは、トピック「 コマンドの順序 」を参照してください。
META BINARY
コマンドのシンタックスは、 「メタ分析 2 値」 ダイアログから生成できます。
リリース履歴
リリース 28.0
- コマンドが導入されました。
例
META BINARY
/DATA TREATMENT=SUCCESS(var_name) FAILURE(var_name) CONTROL=SUCCESS(var_name) FAILURE(reside)
ID=var_name ESTYPE=LOGOR
/CRITERIA CILEVEL=95 SCOPE=AVAILABLE CLASSMISSING=EXCLUDE MAXITER=100 MAXSTEP=100
CONVERGENCE=0.000001 ZEROCOUNTS=ONLY ADD=0.5 DROP=TRUE
/INFERENCE MODEL=RANDOM ESTIMATE=REML ADJUSTSE=NONE.