¡@
|
¦h¼h¦¸¼Ò«¬(HLM)¤Î«½Æ´ú¶q¡G¨Ï¥ÎSTaTa¡]ªþ¥úºÐ¡^
|
|
|
§@¡@¡@ªÌ¢¬
|
±i²Ð¾±
|
¥Xª©ªÀ§O¢¬
|
¤«n
|
®Ñ¡@¡@¨t¢¬
|
¬ã¨s&¤èªk |
¥Xª©¤é´Á¢¬
|
2018/01/12
(1ª© 1¨ê)
|
|
I
S B N ¢¬
|
978-957-11-9506-3
|
®Ñ¡@¡@¸¹¢¬
|
1H0P
|
¶¡@¡@¼Æ¢¬
|
1040
|
¶}¡@¡@¼Æ¢¬
|
16K
|
©w¡@¡@»ù¢¬
|
1050
|
±Ð¾Ç¸ê·½¢¬
|
§ë¼v¤ù((¥~¥[))
|
|
|
•¥þ¤è¦ì¸ÑªR¦h¼h¦¸¼Ò«¬¡]HLM¡^¡AÀ±¸É³æ¼h¦¸°jÂkªº¤£¨¬¡AÅý±z¯à©âµ·éõ¦a¾Ç²ß¡C •¥»®Ñ´£¨Ñ¦h¼h¦¸¼Ò«¬½d¨Ò¸ê®ÆÀÉ¡B²Îpµ{¦¡¾Þ§@¤è¦¡¡A¨Ã±j½Õ²Îp³nÅé³øªí¸ÑÄÀ¡A¨ÏŪªÌ§ó¥[ÁA¸Ñ¨ä¤¤¶ø§®¡C •²`«×²z¸Ñ¥\¯àÃe¤jªºSTaTa¨t²Î¡A¨ä«K§Q©Ê»·¶W¹LR¡BMLWin¡BMplus¡BSPSS¡BSASµ¥³nÅé¡C •¥»®Ñ¾A¦X¥Î©óªÀ·|¬ì¾Ç¡B¥Íª«Âå¾Ç¡B°]¸g¬ã¨sµ¥¹ê¥Î¤u¨ã®Ñ¡A¨Ã¯à¼sªxÀ³¥Î¨ìºÞ²z¡B±Ð¨|¡B¤HºÞ¡BÂåÃÄ°·±dµ¥»â°ì¡C •ÀH®ÑªþÃظê®ÆÀÉ¥úºÐ¡C
STaTa¬O¦aªí³Ì±j²Îp³nÅé¡A¾A¦X¾Ç¥Í¡B¾ÇªÌ»P¹ê°È¤Hû¡I
¬ì¾Ç¬O¤@ªù²Ö¿n¦Ó¦¨ªº¾Ç°Ý¡A²Îp¤S¬O¬ì¾Çªº¹êÃÒ°ò¦¡A·í¤µ¬ì§Þ¹q¸£ªº²Îppºâ¯à¤O¤é·s¤ë²§¡A¯Âºé·§©À©Êªº²z½×¹L©óÀRºA¡A¸ÑÄÀ¹ê»Ú°ÝÃD¥H¤ÎÀ³¥Î¯à¤O¦³¡C¦h¼h¦¸¼Ò«¬(HLM)¶i¦Ó¿³°_¡A¦¨¬°·í¥N²Õ´»PºÞ²z»â°ìªº«n¬ã¨s¨å½d¤§¤@¡ISTaTa³o®M¥\¯àÃe¤jªº²Îp³nÅé¤]¦]¦Ó²£¥Í¡A¥»®Ñ´£¨Ñ½d¨Ò¸ê®Æ®w»P²Îpµ{¦¡¾Þ§@¤è¦¡¡A¦³§U©óÁA¸Ñ¦h¼h¦¸°jÂk¤ÀªR¥Dn¼Ò«¬ªº¾Þ§@µ{§Ç¡A¤º¤åµ²¦X¡u²z½×¡B¤èªk¡B²Îp¡v¡AÅý±z¯à§ó¦³²`«×ªº²z¸Ñ¤°»ò¬O¦h¼h¦¸¼Ò«¬¥H¤ÎSTaTa¡C
|
¡°±ÀÂË¤å ¯Âºé·§©À©Êªº²z½×ÅãµM¦b¹ê»Ú°ÝÃDªº¸ÑÄÀ»PÀ³¥Î¤W¬O¦³ªº¡C¦Ó¯Ê¥F²z½×°ò¦ªº²Îp¤ÀªR«hµLªk´£¨Ñ¤@Ó¡u¦]ªGÃö«Y¡vªº»¡©ú¡C¦P¼Ë¡A¨ä»¡ªA¤O¤]¬O¦³ªº¡A©Î¬Æ¦Ü¥i¯à¬O¿ù»~ªº¡C¥[¤W²Õ´»PºÞ²z¬ã¨sªº¸ê®Æ¦h¯A¤Î¶¥¼h¯S©Ê¡A¦]¦¹¦h¼h¦¸¼Ò«¬¬O·í¥N²Õ´»PºÞ²z»â°ìªº«n¬ã¨s¨å½d¤§¤@¡C¦]¦Ó¿³°_¦h¼h¦¸¼Ò«¬¡]HLM¡^¨ÓÀ±¸É³æ¼h¦¸°jÂkªº¤£¨¬¡C¨´¤µ¡A¦h¼h¦¸»P¼h¯Å½u©Ê¼Ò«¬¼sªxÀ³¥Î¨ìºÞ²z¡B±Ð¨|¡B¤HºÞ¡BÂåÃÄ°·±dµ¥»â°ì¡C ¤H¨Ã«D©t¥ßªºÓÅé¡A¦Ó¬O¾ãÓªÀ·|¤¤ªº¤@û¡A¨Ò¦p¡A¾Ç¥Í¼h¦¸ªº¸ê®Æ±_®M©ó°ª¤@¼hªº¤ÀªR³æ¦ì¡]¦p¯Z¯Å©Î¾Ç®Õ¡^¤§¤º¡A¦b¦P¤@Ó°ª¶¥¤ÀªR³æ¦ì¤UªºÓÅé·|¦]¬°¬Û¦üªº¯S½è¡A§í©Î¨ü¨ì¦@¨ÉªºÀô¹Ò¯ßµ¸©Ò¼vÅT¡A³y¦¨Ó¤H¼h¦¸¸ê®Æ¶¡¨ã¦³¬Û¨Ìªº©Ê½è¡A¥ç§Y¦s¦bµÛ²Õ¤ºÆ[¹î¸ê®Æ¤£¿W¥ßªº²{¶H¡C¥Ñ¦¹¥iª¾¡AÓÅ骺¦æ¬°©Î¤ÏÀ³¤£¶È·|¨ü¨ì¦Û¨¯S©Êªº¼vÅT¡A¤]·|¨ü¨ì¨ä©Ò³BªºÀô¹Ò¯ßµ¸©Ò¼vÅT¡C¦]¦¹±_ª¬¡]nested¡^¸ê®Æ¤ÀªR¤§¦h¼h¦¸¼Ò«¬´N¦¹¬y¦æ°_¨Ó¡C ¬ì¾Ç¬O¤@ªù²Ö¿nªº¾Ç²ß½Òµ{¡A²Îp¤S¬O¬ì¾Ç¤§¹êÃÒ°ò¦¡A¥[¤W¹q¸£²Îppºâ¯à¤O¤é·s¤ë¯q¡C¶i¦Ó¹³STaTa³oºØÃe¤j¥\¯à¤§²Îp³nÅ骺½Ï¥Í¡ASTaTa¥\¯à¤Î«K§Q©Ê³£»·¶W¶VR¡BMLWin¡BMplus¡BSPSS¡BSAS¡BAMOS¤ÎEviewsµ¥³nÅé¡CSTaTaÀu¨q²Îp¥\¯à¬O¦aªí³Ì±jªº³nÅé¡CSTaTa¤w¥i³B²z¡G(1)¾îÂ_±¬ã¨s³]p¡BÁa³e±¬ã¨s³]p¡BÁa¾î±(pandel-data)¬ã¨s³]p¡C(2)³æ¤@°jÂk¤èµ{¦¡¡B°jÂkÁp¥ß¤èµ{¦¡¡C(3)ºô¯¸´£¨ÑÃe¤j½u¤W¨D§U¡B¥~±¾«ü¥O¡]*.ado¡^¨Óº¡¨¬¦U¦æ¦U·~ªºÓ§O»Ý¨D¡C(4)³æÅܶq¡B¦hÅܶq®É¶¡§Ç¦C¥ç¥i³B²z¡C(5)¦h¼h¦¸vs.³æ¼h¦¸°jÂk¤§²Îp¤ÀªR¡K¡Kµ¥³£¦³³Ì·s³B²z¤èªk¡A¥ú¬O¦h¼h¦¸½u©Ê²V¦X¼Ò«¬´N¦³15ºØ¥H¤WÃþ«¬¥i¨Ñ¿ï¾Ü¡C STaTa¦P®É´£¨Ñ²³¦h¡]¤º«Øvs.¥~±¾¡^«ü¥O¡A´X¥G§{¶¡±Ð¬ì®Ñ§A¬Ý¨ìªº²Îp¤ÀªR¡A¥¦³£¥i¸Ñ¨M¡C¦¹¥~¡ASTaTa¬°¤F´î§C¹q¸£¨Ï¥ÎªÌ¹ïµ{¦¡³]pªº¼~¼{¡A¥¦¥ç´£¨ÑMenu¿ï¾Üªí¤§¹ïÀ³µøµ¡¡AÅý¹³SPSS¡BSAS¡BHLM¨º¼Ë¯à»´ÃP¾Þ§@Menu¡A¨Ó¶i¦æ²Îp¤ÀªR¡Cªp¥B¡ASTaTa¦³¯à¤O³B²z¹³big-data³oÃþ¥¨¤j¸ê®Æ®w¡C ¦³Å²©óSTaTa¤ÀªR¥\¯àÃe¤j¡A¬G§@ªÌ±N¼¶¼g¤@¨t¦CªºSTaTaªº®Ñ¡A¥]¬A¡G 7.¦h¼h¦¸¼Ò«¬¡]¦@15ºØ¦h¼h¦¸¼Ò«¬¤§Ãþ«¬¥i¿ï¾Ü¡^¡C ªÀ·|¬ì¾Ç¡B¥Íª«Âå¾Ç¡B°]ª÷µ¥»â°ì¡A¨ä²Îp¾Ç«Y±Ä¥Î²Îp¾Ç¡B¹BÄw¾Ç¡B¸gÀÙ¡B¼Æ¾Çµ¥»â°ì¤§©w¶q¤èªk¡CªÀ·|¬ì¾Ç¤Î¦ÛµM¬ì¾Ç¤G¤j»â°ì¤¤¦UӾǬì¡A¥¦Ì¦³³\¦h¦@³q¤§¬ã¨s³]p¤Î²Îp¤ÀªRªk¡A³£»P§@ªÌ¦b¤«n¥Xª©STaTa¤@§Ç¦C®Ñ¦W¦³Ãö¡A¥]¬A¡G
¤@¡B¡mSTaTa»P°ªµ¥²Îp¤ÀªRªºÀ³¥Î¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡G´yz©Ê²Îp¡B¼Ë¥»¼Æªºµû¦ô¡BÅܲ§¼Æ¤ÀªR¡B¬ÛÃö¡B°jÂk«Ø¼Ò¤Î¶EÂ_¡B«½Æ´ú¶q¡K¡K¡C
¤G¡B¡m¡uSTaTa¦bµ²ºc¤èµ{¼Ò«¬¤Î¸ÕÃD¤ÏÀ³²z½×¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡G¸ô®|¤ÀªR¡Bµ²ºc¤èµ{¼Ò«¬¡B´ú¶q¤u¨ãªº«H®Ä«×¤ÀªR¡B¦]¯À¤ÀªR¡K¡K¡C
¤T¡B¡mSTaTa¦b¥Íª«Âå¾Ç²Îp¤ÀªR¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡GÃþ§O¸ê®Æ¤ÀªR(µL¥À¼Æ²Îp)¡Blogistic°jÂk¡B¦s¬¡¤ÀªR¡B¬y¦æ¯f¾Ç¡B°t¹ï»P«D°t¹ï¯f¨Ò¹ï·Ó¬ã¨s¸ê®Æ¡B²±¦æ²v¡Bµo¥Í²v¡B¬Û¹ï¦MÀI²v¤ñ¡B³Ó¥X¤ñ¡]Odds Ratio¡^ªºpºâ¡B¿zÀˤu¨ã»PROC¦±½u¡B¤u¨ãÅܼơ]2SLS¡^¡K¡KCox¤ñ¨Ò¦MÀI¼Ò«¬¡BKaplan-Meier¦s¬¡¼Ò«¬¡B¯Ü®z©Ê¤§Cox¼Ò«¬¡B°Ñ¼Æ¦s¬¡¤ÀªR¦³¤»ºØ¼Ò«¬¡B¥[³t¥¢±Ñ®É¶¡¼Ò«¬¡Bpanel-data¦s¬¡¼Ò«¬¡B¦h¼h¦¸¦s¬¡¼Ò«¬¡K¡K¡C
¥|¡B¡mMeta²Îp¤ÀªR¹ê§@¡G¨Ï¥ÎExcel»PCMAµ{¦¡¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡G²Î¦X¤ÀªR¡]meta-analysis¡^¡B³Ó¥X¤ñ¡]Odds Ratio¡^¡B·ÀI¤ñ¡B4ºØ¦³¦W®ÄªG¶q¡]ES¡^¤½¦¡¤§³æ¦ìÅÜ´«µ¥¡C
¤¡B¡mPanel-data°jÂk¼Ò«¬¡GSTaTa¦b¼s¸q®É¶¡§Ç¦CªºÀ³¥Î¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡G¦h¼h¦¸¼Ò«¬¡BGEE¡B¤u¨ãÅܼơ]2SLS¡^¡B°ÊºA¼Ò«¬¡K¡K¡C
¤»¡B¡mÁ`Åé¸gÀÙ»P°]°Èª÷¿Ä¡GSTaTa®É¶¡§Ç¦C¤ÀªR¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡G»~®t²§½è©Ê¡B°ÊºA¼Ò«¬¡B§Ç¦C¬ÛÃö¡B®É¶¡§Ç¦C¤ÀªR¡BVAR¡B¦@¾ã¦X¡K¡Kµ¥¡C
¤C¡B¡m¦h¼h¦¸¼Ò«¬¡]HLM¡^¤Î«½Æ´ú¶q¡G¨Ï¥ÎSTaTa¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡G½u©Ê¦h¼h¦¸¼Ò«¬¡Bvs.Â÷´²«¬¦h¼h¦¸¼Ò«¬¡Bp¼Æ«¬¦h¼h¦¸¼Ò«¬¡B¦s¬¡¤ÀªR¤§¦h¼h¦¸¼Ò«¬¡B«D½u©Ê¦h¼h¦¸¼Ò«¬¡K¡K¡C
¤K¡B¡m¼Ò½k¦h·Ç«hµû¦ôªk¤Î²Îp¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡GAHP¡BANP¡BTOPSIS¡BFuzzy²z½×¡BFuzzy AHP¡K¡Kµ¥²z½×»P¹ê§@¡C
¤E¡B¡mÅÞ¿è´µ°jÂk¤ÎÂ÷´²¿ï¾Ü¼Ò«¬¡GÀ³¥ÎSTaTa²Îp¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡GÅÞ¿è´µ°jÂk¡Bvs. ¦h¤¸ÅÞ¿è´µ°jÂk¡B°t¹ï¸ê®Æªº±ø¥óLogistic°jÂk¤ÀªR¡BMultinomial Logistic Regression¡B¯S©w¤è®×Rank-ordered logistic°jÂk¡B¹s¿±µÈordered probit regression°jÂk¡B°t¹ï¸ê®Æªº±ø¥óÅÞ¿è´µ°jÂk¡B¯S©w¤è®×conditional logit model¡BÂ÷´²¿ï¾Ü¼Ò«¬¡B¦h¼h¦¸ÅÞ¿è´µ°jÂk¡K¡K¡C
¤Q¡B¡m¦³²V¦X¼Ò«¬FMM:STaTa¥ÎEM algorithm¥ý¼ç¦b¤ÀÃþ¦A¹w´ú¡n¤@®Ñ¡A¸Ó®Ñ¤º®e¥]¬A¡GFMM¡G½u©Ê°jÂk¡BFMM¡G¦¸§Ç°jÂk¡BFMM¡GLogit°jÂk¡BFMM¡G¦h¶µLogit°jÂk¡BFMM¡G¹s¿±µÈ°jÂk¡BFMM¡G°Ñ¼Æ«¬¦s¬¡°jÂk¡K¡Kµ¥²z½×»P¹ê§@¡C
³o¥»®Ñªº¥Ø¼Ð§Y¦b´£¨Ñ¦³¿³½ì¦h¼h¦¸¤ÀªRªº¾Ç¥Í¡B¾ÇªÌ»P¹ê°È¤Hû±Ð¾Ç»P¬ã¨sªº«ü¾É¡C¦b³o¥»®Ñ¥ç´£¨Ñ½d¨Òªº¸ê®ÆÀÉ»P²Îpµ{¦¡¾Þ§@¤è¦¡¡C³o¨Ç°ò¦²z½×»P½d¨Ò¦³§U©ó§ÚÌÁA¸Ñ¦h¼h¦¸¤ÀªR¡Cn²`¤JÁA¸Ñ³o¥»®Ñ¥²¶·n¦³°ò¥»ªº¦h¤¸°jÂk¤èµ{¦¡ªºª¾ÃÑ¡C³o¥»®Ñ·|´£¨ÑSTaTaªº½d¨Ò¡A°w¹ï¦h¼h¦¸°jÂk¤ÀªR¥Dn¼Ò«¬ªº¾Þ§@µ{§Ç¡A¨Ã±j½Õ³o¨Ç²Îp³nÅé©Ò²£¥Íªº³øªí¸ÑÄÀ¡A½Ñ¦pµ²ªGÅܼƬ°³sÄòÅܼƩΪÌÃþ§OÅܼƪº¨â¼h»P¤T¼h¼Ò«¬¡B»²§U©Ê²Îp(¦p°jÂk«Y¼Æ«H¿à°Ï¶¡¡B²Õ¤º¬ÛÃö«Y¼Æµ¥)¡BÀ˵ø¼Ò«¬°²³]µ¥¡C ¦¹¥~¡A¬ã¨sªÌ¦p¦ó¿ï¾Ü¥¿½Tªº²Îp¤èªk¡A¥]¬A¾A·íªº¦ôp»PÀË©w¤èªk¡B»P²Îp·§©Àµ¥¡A³£¬O¹êÃÒ¬ã¨s¤¤«Ü«nªº¤º²[¡A³o¤]¬O¥»®Ñ¼¶¼gªº¥Øªº¤§¤@¡C¬°¤FÅý¬ã¨sªÌ¯à¥¿½T¥Bºë·Ç¨Ï¥Îpanel°jÂk¡A¥»®Ñ¤º¤å¾¨¶qµ²¦X¡u²z½×¡B¤èªk¡B²Îp¡v¡A´Á±æ¯à°÷¹ï²£¾Ç¬É¦³©ß¿j¤Þ¥Éªº®ÄªG¡C ³Ì«á¡A¯S·PÁÂ¥þ³Ç¬ì§Þ¤½¥q¡]http://www.softhome.com.tw¡^¡A´£¨ÑSTaTa³nÅé¡A±ß¾Ç¤~¦³¾÷·|¼¶¼gSTaTa¤@¨t¦Cªº®Ñ¡A¥H¹Å´f¾Ç²ßªÌ¡C
±i²Ð¾± ±i¥ô§{ ±i³Õ¤@ ·q¤W
|
¡½§@ªÌ²¤¶ ±i²Ð¾± ¾Ç¾ú¡G°ê¥ß¬Fªv¤j¾Ç¸ê°TºÞ²z³Õ¤h ²{¥ô¡G°ê¥ß¹ü¤Æ®v¤j±M¥ô±Ð±Â ¸g¾ú¡GP²z§Þ³N±M¥ô°Æ±Ð±Â
¡½¬ã¨s§U²z ±i¥ô§{ °ê¥ß®ü¬v¤j¾Ç°Ó²î¨t
±i³Õ¤@ °ê¥ß¤¤¥¡¤j¾Ç³q°T¤uµ{©Ò
|
Chapter 01 STaTa¬O¦aªí³Ì±j²Îp¡A¾A¦X¦U²£©x¾Ç¬ã 1-1 ²Îp¤ÀªR 1-1-1¡@»{ÃѲÎp 1-1-2¡@²Îp»P¡u¹êÅçªk¡BÆ[¹îªk¡v¤§¹ïÀ³Ãö«Y 1-2 STaTa¥@¤W³Ì±j¤jªº²Îp¥\¯à 1-2-1¡@³æ¼h¦¸¡G³sÄòvs.Ãþ§O¨ÌÅܼưjÂk¤§ºØÃþ 1-2-2¡@STaTa ¦h¼h¦¸²V¦X¼Ò«¬ªº°jÂkºØÃþ 1-2-3¡@STaTa panel-data°jÂkªººØÃþ 1-2-4¡@STaTa¬y¦æ¯f(epidemiologists)¤§¿ï¾Üªí¹ïÀ³ªº«ü¥O 1-2-5¡@STaTa¦s¬¡¤ÀªRªº¿ï¾Üªí¤§¹ïÀ³«ü¥O 1-2-6¡@STaTaÁa³e±¡X®É¶¡§Ç¦C¤§¿ï¾Üªí 1-2-7¡@STaTa¦³²V¦X¼Ò«¬(FMM)¡GEM algorithm¿ï¾Üªí 1-3 STaTa¦w¸Ë³]©w 1-4 ¸ê®Æ¿é¤Jªº¤èªk¡G°Ý¨÷¡BExcel 1-5 SPSS¸ê®ÆÀÉ(*.sav)ÂনSTaTa®æ¦¡ 1-6 SAS®æ¦¡ÂনSTaTa 1-7 R®æ¦¡ÂনSTaTa 1-8 ¥~±¾ªº©R¥OÀÉado¡GSTaTa¥~±¾ªºPackage 1-9 »{ÃÑ¡u¦h¼h¦¸¼Ò«¬¡v 1-10 Ãþ»E(clustered)¡þ±_ª¬¸ê®Æ¤ÀªR¡ASTaTa°jÂk¦³16ºØ¦ôpªk 1-11 ¤j¼Æ¾Ú(big data)»PSTaTa¸ê®ÆÀɤ§¶¡ªº®æ¦¡¥i¤¬³q Chapter 02 ¦h¼h¦¸¤ÀªRªk¡GHLM 2-1 ¦h¼h¦¸¼Ò«¬(¶¥¼h½u©Ê¼Ò«¬HLM)ªº¿³°_ 2-1-1¡@¦h¼h¦¸¼Ò«¬(¶¥¼h½u©Ê¼Ò«¬HLM)ªº¿³°_ 2-1-2¡@³æ¼h¦¸¡G¦h¤¸°jÂk¤ÀªR(OLS)¤§«ÂI¾ã²z 2-2 ¤°»ò¬O¦h¼h¦¸¤ÀªRªk¡H 2-2-1¡@¶¥¼h½u©Ê¼Ò«¬(HLM)¤§¥Ñ¨Ó 2-2-2¡@¦h¼h¦¸¼Ò«¬¤§«n©Ê 2-2-3¡@¶Ç²Î³æ¼h¦¸OLS¤ÀªR±_ª¬(nested)¸ê®Æ¡A·|¥X¤°»ò°ÝÃD¡H 2-2-4¡@¯ßµ¸ÅܼÆ(contextual variables) vs. Á`ÅéÅÜ¼Æ 2-3 ¦h¼h¦¸¤ÀªR¤§¼Ò«¬¬É©w 2-3-1¡@¦h¼h¦¸¼Ò«¬¤§¥Ü·N¹Ï 2-3-2¡@¦h¼h¦¸¼Ò«¬¤§°²©w(assumption) 2-3-3¡@ÀH¾÷ºI¶Zvs.ÀH¾÷±×²v¤§¥|ºØÃö«Y 2-3-4a¡@ÀH¾÷«Y¼Æ¼Ò«¬¤§¤TºØ³]©w(random coefficient modeling) 2-3-4b¡@Âù¦]¤lÀH¾÷«Y¼Æ¤§¤TºØ³]©w¸Ñ»¡144 2-3-5¡@¦h¼h¦¸¸ê®Æµ²ºc¡G¥´î(centering)§YÂ÷®t¤À¼Æ(deviated scores) 2-4 ¼Ò«¬³]©w¡þ«Øºcªº¨BÆJ 2-4-1¡@¼Ò«¬³]©wªº¨BÆJ 2-4-2¡@¦p¦ó´£¤É¦h¼h¦¸¤ÀªRªkªºÄYÂÔ©Ê 2-5 ÅܼƤ¤¤ß¤Æ(centering)¡B¥æ¤¬§@¥Î¶µ(Z¡ÑA)¨ã¦h¤¸¦@½u©ÊºÃ¼{ 2-5-1¡@¬°¦óÁ`¥´î(grand-mean centering)¥i§JªA¦h¤¸¦@½u©Ê¤§°ÝÃD 2-5-2¡@¥æ¤¬§@¥Î¶µ(Z*A)·|¾É¦Ü¦h¤¸¦@½u©Ê¤§ÄY«°ÝÃD¡G¤ßŦ¬ì 2-5-3¡@ÅܼƤ¤¤ß¤Æ(centering variables)¡þ¥´î 2-5-4¡@¤¤¤ß¤Æ(centering)¡þ¥´îªº®É¾÷ 2-5-5¡@¤¤¤ß¤Æ(centering)ªºÃþ§O 2-6 ½u©Ê²V¦X¼Ò«¬¡G¦h¼h¦¸¤ÀªR¤Jªù(mixed, xtmixed«ü¥O) 2-6-1¡@½u©Ê²V¦X¼Ò«¬(linear mixed model)¤§¤èµ{¦¡ 2-6-2¡@½d¨Ò1¡G¨å«¬¤§ÀH¾÷«Y¼Æ¼Ò«¬(slopes and intercepts as outcomes) 2-6-3¡@½d¨Ò2¡G¦Ò¶q¦U¤p¸s²Õ¤§»~®tµ²ºc(slopes and intercepts as outcomes with variance-covariance structure of the random effects) 2-7 ¦p¦ó±N¦h¼h¼Ò«¬Âন²V¦X¼Ò«¬(ml2mixed¥~±¾«ü¥O) 2-8 ¦]ªGÃö«Yªº²Ä¤TªÌ¡G½Õ¸`¡þ¤zÂZÅܼÆ(moderator)¡B¤¤¤¶ÅÜ¼Æ 2-8-1¡@²Õ´¬ã¨sªº¤¤¤¶ÀË©w¤§½t°_ 2-8-2¡@¤¤¤¶ÅܼÆ(ª½±µ®ÄªG¡B¶¡±µ®ÄªG)¡Ú½Õ¸`ÅܼÆ(¥æ¤¬§@¥Î®ÄªG) 2-8-2a¡@¤¤¤¶ÅܼÆ(mediator variable) 2-8-2b¡@¤¤¤¶ÅܼÆ(mediator variable)¦s¦b»P§_ªº¥|ºØÀË©wªk 2-8-3¡@½Õ¸`ÅܼÆ(moderator variable)¡A¤SºÙ¤zÂZÅÜ¼Æ 2-8-4¡@½Õ¸`¦¡¤¤¤¶®ÄªG(moderated mediation effect) 2-8-5¡@¦h¼h¦¸¤¤¤¶®ÄªG¡GSTaTa¹ê§@(ml_mediation¡Bxtmixed«ü¥O) 2-8-5a¡@¦h¼h¦¸¤¤¤¶®ÄªG¡GSTaTa¤èªk¤@(ml_mediation«ü¥O) 2-8-5b¡@Âù¼h¦¸¤¤¤¶®ÄªG¡GSTaTa¤èªk¤G(xtmixed¡Bmixed«ü¥O) 2-8-6¡@Sobel-Goodman¤¤¤¶ÀË©wªk(¥ýsgmediation¦Aml_mediation«ü¥O) Chapter 03 ³æ¼hvs.Âù¼h¦¸¼Ò«¬¡GµL¥æ¤¬§@¥Î¶µ´NµL¶·¤¤¤ß¤Æ 3-1 ¦h¼h¦¸¼Ò«¬¤§«ÂI¸É¥R 3-1-1¡@¤À¼hÀH¾÷©â¼Ë 3-1-2¡@Panel-data°jÂk¼Ò«¬¤§«ÂI¾ã²z 3-2 ³æ¼hvs.Âù¼h¡G«½Æ´ú¶qªº²V¦X®ÄªG¼Ò«¬(mixed effect model for repeated measure) 3-2-1¡@ANOVA¤ÎµL¥À¼Æ²Îp¤§¤ÀªR¬yµ{¹Ï 3-2-2¡@«½Æ´ú¶qANOVA¤§FÀË©w¤½¦¡ 3-2-3¡@³æ¼h¦¸¡G¤G¦]¤l²V¦X³]pANOVA (anova¡Bcontrast¡Bmargin¡Bmarginsplot«ü¥O) 3-2-4¡@«½Æ´ú¶qANOVA¤§¥Dn®ÄªG¡þ³æ¯Â¥Dn®ÄªGÀË©w(Âù¼hxtmixed©Îmixed vs.³æ¼hanova«ü¥O) 3-2-5¡@Âù¼h¦¸¡G¤G¦]¤l²V¦X³]pANOVA (mixed©Îxtmixed«ü¥O) 3-3 ¼Ä¹ï¼Ò«¬Ì¨º¤@Ó¸ûÀu©O¡H¥ÎIC¸ê°T·Ç«h(mixed, xtmixed«ü¥O) 3-3-1¡@°»´ú¨âӼĹï¼Ò«¬¡A¾A°t«ü¼Ð¦³7ºØ 3-3-2¡@±Æ¦C²Õ¦X¤@¡G3ºØ¼Ä¹ï¼Ò«¬(mixed, xtmixed«ü¥O) 3-3-3¡@±Æ¦C²Õ¦X¤G¡G10ºØ¼Ä¹ï¼Ò«¬(mixed, xtmixed«ü¥O) 3-3-4¡@±Æ¦C²Õ¦X¤T¡G©ú¬P¾Ç®Õ¯uªº¤ñ¸û¦n¶Ü¡G4ºØ¼Ä¹ï¼Ò«¬ 3-3-5¡@±Æ¦C²Õ¦X¥|¡GµLvs.¦³¥æ¤¬§@¥Î¶µ¡A¨ºÓ¼Ò«¬¦n©O¡H(mixed, xtmixed«ü¥O) Chapter 04 ¦h¼h¦¸¼Ò«¬¤§¤èµ{¦¡¸Ñ»¡¡G¦³(Z¡ÑX)¥æ¤¬§@¥Î¶µ´N¶·¤¤¤ß¤Æ 4-1 ¦h¼h¦¸¼Ò«¬¤§¤èµ{¦¡¸Ñ»¡¡G¼vÅT¦í¦v©Ð»ù¤§ÓÅé¼h¤Î¸s²Õ¼h 4-1-1¡@Step 1³]©w(¼Ò«¬1)¡G¹s¼Ò«¬(null model) 4-1-2¡@Step 2³]©w(¼Ò«¬2)¡G¥§¡¼Æ¬°µ²ªGªº°jÂk¼Ò«¬(means-as-outcomes regression) 4-1-3¡@Step 3³]©w(¼Ò«¬3)¡GLevel-1¨ã©T©w®ÄªG¤§ÀH¾÷ºI¶Z¼Ò«¬ 4-1-4¡@Step 4³]©w(¼Ò«¬4):ÀH¾÷«Y¼Æ(random coefficients)°jÂk¼Ò«¬ 4-1-5¡@Step 5³]©w(¼Ò«¬5):ºI¶Z»P±×²v¬°µ²ªGªº°jÂk(¥æ¤¬§@¥Î) Chapter 05 ¦h¼h¦¸¼Ò«¬¤§STaTa¹ê§@¤Î¸Ñ»¡(·sª©mixed, ª©xtmixed«ü¥O) 5-1 ¤»¨BÆJ¨Ó¬D¿ï³Ì¨Î¦h¼h¦¸¼Ò«¬(§YHLM)¡H¥ÎIC·Ç«h¨Ó§PÂ_ 5-1-0¡@¼Ë¥»¸ê®ÆÀÉ 5-1-1¡@Step 1¡G¹s¼Ò«¬(intercept-only-model, unconditional model) 5-1-2¡@Step 2¡GLevel-1³æ¦]¤l¤§ÀH¾÷ºI¶Z¼Ò«¬(µLÀH¾÷±×²vu1j) 5-1-3¡@Step 3¡GLevel-1³æ¦]¤l¤§ÀH¾÷ºI¶Z¥BÀH¾÷±×²v¼Ò«¬(slopes and intercepts as outcomes) 5-1-4¡@Step 4¡GLevel-1Âù¦]¤l¤§ÀH¾÷±×²v¼Ò«¬(slopes and intercepts as outcomes) 5-1-5¡@Step 5¡GLevel-2³æ¦]¤l¤ÎLevel-1 Âù¦]¤l¤§ÀH¾÷¼Ò«¬(µL¥æ¤¬§@¥Î) 5-1-6¡@Step 6¡GLevel-2³æ¦]¤l¤ÎLevel-1 Âù¦]¤l¤§ÀH¾÷¼Ò«¬(¦³¥æ¤¬§@¥Î) 5-2 ¦h¼h¦¸¼Ò«¬¤§STaTa½m²ßÃD(·sª©mixed«ü¥O¡Aª©xtmixed«ü¥O) Chapter 06 ³æ¼h¦¸vs.¦h¼h¦¸¡GÂ÷´²«¬¨ÌÅܼƤ§Poisson°jÂk 6-1 ³æ¼h¦¸Count¨ÌÅܼơGZero-inflated Poisson°jÂk vs. negative binomial°jÂk 6-1-1¡@Poisson¤À°t 6-1-2¡@t¤G¶µ¤À°t(negative binomial distribution) 6-1-3¡@¹s¿±º¦(Zero-inflated) Poisson¤À°t 6-2 ³æ¼h¦¸¡GZero-inflated Poisson°jÂkvs.t¤G¶µ°jÂk(zip¡Bzinb«ü¥O) 6-3 ¤T¼h¦¸¡GPoisson°jÂk(mepoisson ©Îxtmepoisson«ü¥O) 6-3-1¡@¦h¼h¦¸Poisson¼Ò«¬ 6-3-2¡@¤T¼h¦¸¡GPoisson°jÂk(mepoisson©Îxtmepoisson«ü¥O) 6-4¡@½m²ßÃD¡GÂù¼hÀH¾÷ºI¶Z¼Ò«¬¤§Poisson°jÂk(mepoisson«ü¥O) Chapter 07 ³æ¼h¦¸vs.Âù¼h¦¸¡G¤G¤¸¨ÌÅܼƤ§Logistic°jÂk 7-1 Logistic°jÂk¤§ì²z 7-1-1¡@³Óºâ¤ñ(OR) 7-1-1a¡@³Óºâ¤ñ(odds ratio)¤§·N¸q 7-1-1b¡@odds ratio¤§STaTa¹ê§@ 7-2 ³æ¼h¦¸¡GLogistic°jÂk(logit«ü¥O) 7-2-1¡@Logit¼Ò«¬¤§¸Ñ»¡ 7-2-2¡@³æ¼h¦¸¡G¤G¤¸¨ÌÅܼƤ§¼Ò«¬¡GLogistic°jÂk¤§¹ê¨Ò 7-3 ½d¨Ò¡G¤T¼h¦¸:Logistic°jÂk(melogit©Îxtmelogit«ü¥O) 7-4 ½m²ßÃD¡GÂù¼h¦¸Logistic°jÂk(melogit«ü¥O) Chapter 08 ½d¨Ò¡GÂù¼h¦¸vs.¤T¼h¦¸¡G½u©Ê¦h¼h¦¸¼Ò«¬ 8-1 Âù¼h¦¸²V¦X(multilevel mixed)¼Ò«¬ 8-1-1¡@Âù¼h¦¸¡Gmixed©Îmultilevel©Îhierarchical model (xtmixed«ü¥O) 8-1-2¡@Âù¼h¦¸¡G¦h¼h¦¸¦¨ªø¼Ò«¬(xtmixed«ü¥O) 8-1-3¡@Âù¼h¦¸¡G¦h¼hÀH¾÷ºI¶Z/ÀH¾÷±×²v¼Ò«¬(xtmixed«ü¥O) 8-1-4¡@Âù¼h¦¸¡G²§½è©Ê»~®t¤§ÀH¾÷ºI¶Z©Î²V¦X®ÄªG¼Ò«¬(xtmixed«ü¥O) 8-1-5¡@Âù¼h¦¸¡GÂù¼h¦¸²V¦XLogistic°jÂk(xtmelogit«ü¥O) 8-1-6¡@Âù¼h¦¸¡G¼ç¦b¦¨ªø¦±½u(xtmixed+ nlcom«ü¥O) 8-2 ¤T¼h¦¸²V¦X(multilevel mixed)¼Ò«¬ 8-2-1¡@¤T¼h¦¸¯ßµ¸¼Ò«¬¡G½u©Ê²V¦X°jÂk(xtmixed«ü¥O) 8-2-2¡@¤T¼h¦¸¡GÀH¾÷ºI¶Z/ÀH¾÷±×²v¼Ò«¬(xtmixed«ü¥O) Chapter 09 ³æ¼hvs.Âù¼h¡GCox¦s¬¡¤ÀªR¡GÁ{§É³Ì«n²Îpªk 9-1 ¦s¬¡¤ÀªR(survival analysis)¤¶²Ð 9-1-1¡@¦s¬¡¤ÀªR¤§©w¸q 9-1-2¡@¬°¦ó¦s¬¡¤ÀªR¬OÁ{§É¬ã¨s³Ì«nªº²Îpªk¡H 9-1-3¡@¦s¬¡¤ÀªR¤§¤TºØ¬ã¨s¥Ø¼Ð 9-1-4¡@¦s¬¡¤ÀªR¤§¬ã¨sijÃD 9-1-5¡@³]¸ê®Æ(censored data) 9-1-6¡@¦s¬¡®É¶¡T¤§¾÷²v¨ç¼Æ 9-1-7¡@Cox¦s¬¡¤ÀªRvs. Logit¼Ò«¬/probit¼Ò«¬ªº®t²§ 9-2 STaTa¦s¬¡¤ÀªR¡þø¹Ïªí¤§¹ïÀ³«ü¥O¡B·s¼W²Îp¥\¯à 9-3 ¦s¬¡¤ÀªR½d¨Ò¡G°£¯ó¦³§U¥®]¦s¬¡²v¶Ü¡H 9-3-1¡@¥Í©Rªí(life table) 9-3-2¡@¦s¬¡¤ÀªR½d¨Ò[¨Ì§Ç(estat phtest¡Bsts graph¡Bltable©Îsts list¡Bstci¡Bstmh¡Bstcox«ü¥O)] 9-4 Cox¤ñ¨Ò¦MÀI¼Ò«¬(proportional hazards model)(stcox«ü¥O) 9-4-1¡@f(t)¾÷²v±K«×¨ç¼Æ¡BS(t)¦s¬¡¨ç¼Æ¡Bh(t)¦MÀI¨ç¼Æ¡BH(t)²Ö¿n¦MÀI¨ç¼Æ 9-4-2¡@Cox¤ñ¨Ò¦MÀI¼Ò«¬¤§°jÂk¦¡¸Ñ»¡ 9-4-3¡@¦MÀI¨ç¼Æªº¦ôp(hazard function) 9-4-4¡@Cox¤ñ¨Ò¦MÀI¼Ò«¬¤§¾A°t«×ÀË©w 9-5 ³æ¼h¦¸¡G¨ã¯Ü®z©ÊCox¼Ò«¬(Cox regression with shared frailty) 9-5-1¡@¯Ü®z©Ê¤§Cox¼Ò«¬¡G¡ustcox, shared(¯Ü®zÅܼÆ)¡v«ü¥O 9-6 ±a°¾ºA¤§¨ÌÅܼơG°Ñ¼Æ¦s¬¡¤ÀªR(streg«ü¥O) 9-6-1¡@¯Ü®z©Ê(frailty)¼Ò«¬ 9-6-2¡@¥[³t¥¢±Ñ®É¶¡(accelerated failure time)¼Ò«¬ 9-7 Âù¼h¦¸¡Gpanel-data°Ñ¼Æ¦s¬¡¼Ò«¬[xtstreg, shared(panelÅܼÆ)«ü¥O] 9-7-1¡@°lÂܸê®Æ(panel-data) 9-7-2¡@°lÂܸê®Æ(panel-data)¦s¬¡¤ÀªR[xtstreg, shared(panelÅܼÆ)«ü¥O] 9-8 ¦h¼h¦¸¡G°Ñ¼Æ¦s¬¡¼Ò«¬(mestreg¡B¡usttocc clogit¡v«ü¥O) 9-8-1¡@multilevel¦s¬¡¼Ò«¬ 9-8-2¡@¦h¼h¦¸°Ñ¼Æ¦s¬¡¼Ò«¬(mestreg¡K ||¤À¼hÅܼÆ) 9-9 ±_ª¬«¬¯f¨Ò¡X¹ï·Ó¬ã¨sªk(nested case-control) (¥ýsttocc¦Aclogit«ü¥O) 9-10 ½m²ßÃD¡G¤T¼h¦¸¤§¹ï¼Æ±`ºA¦s¬¡¼Ò«¬(mestreg«ü¥O) Chapter 10 «D½u©Ê¡G¦h¼h¦¸²V¦X®ÄªG¼Ò«¬(menl«ü¥O) 10-1 «D½u©ÊÁa³e±¸ê®Æ¡GÀH¾÷ºI¶Z¤§¦h¼h¦¸¼Ò«¬¡X¿W¨¤Ã~ 10-2 «D½u©Ê½d¨Ò¡G»~®tµL¦@Åܵ²ºc¤§Âù¼h¼Ò«¬¡X¾ï¤l¾ð 10-3 «D½u©Ê¦h¼h¦¸¼Ò«¬¡G¸s²Õ¤º¤§»~®t¬ÛÃöµ²ºc¡Xovary 10-4 «D½u©Ê¡G¤T¼h¦¸¼Ò«¬¡X¦å¿}(blood glucose) 10-5 ´Ý®t¦³¦@ÅܼƵ²ºc¡GÃÄ¥N°Ê¤O¾Ç«Ø¼Ò¡XPharmacokinetic(PK) model Chapter 11 »~®tÅܲ§2¨ã²§½è©Ê(xtgls«ü¥O¬°¥D¬y) 11-1 ´Ý®t¤§Åܲ§¼Æ 11-1-1¡@»~®tÅܲ§ ªºÆ[©À 11-1-2¡@»~®tÅܲ§ ªº°»´úªk 11-2 ³æ¼h¦¸¡G°»´ú»~®t¤§²§½è©Ê(heteroskedasticity) 11-2-1¡@¾îÂ_±OLS°jÂk¡G´Ý®t²§½è©Ê¶EÂ_(hettest«ü¥O) 11-2-2¡@´Ý®t²§½èªº§ïµ½¡GOLS§ï¦¨robust°jÂk 11-2-3¡@¾îÂ_±¤§»~®t²§½è©Ê¡G»Ýln()ÅܼÆÅÜ´«(¥ýreg¦Awhitetst«ü¥O) 11-2-4¡@Áa³e±¤§»~®t²§½è©Ê(¥ýreg¦Abpagan«ü¥O) 11-3 ¦h¼h¦¸¡G¨ã²§½è©Ê»~®t¤§ÀH¾÷ºI¶Z¡þ²V¦X¼Ò«¬(xtmixed¡Bmixed«ü¥O) 11-3-1¡@½d¨Ò1¡G¨D¦U²Õ¤§»~®t²§½è(heteroskedastic errors by group) 11-3-2¡@½d¨Ò2¡GÁa³e±¤§¦¨ªø¦±½u¼Ò«¬(«½Æ´ú¶q5¦¸) 11-3-3¡@½d¨Ò3¡G»~®tÅܲ§¼Æ2error¨ã²§½è©Ê °Ñ¦Ò¤åÄm
|
§ë¼v¤ù((¥~¥[))
Yn¯Á¨ú¥¼ÀH®Ñªþ°e(¥~¥[)¥B¥¼©ó¦¹´£¨Ñ¤U¸üªº±Ð¾Ç¸ê·½¡A½Ð¸Ô¬¢·~°È¤Hû(02-27055066#824)(¶È´£¨Ñ±Ð®v¨Ï¥Î)
|
|
|
2-1¦h¼h¦¸¼Ò«¬(¶¥¼h½u©Ê¼Ò«¬HLM)ªº¿³°_ ¦h¼h¦¸¼Ò«¬(multilevel models,MLM)¡A¤SºÙ¡G¶¥¼h½u½u¼Ò«¬(hierarchical linear models, HLM)¡B±_ª¬¸ê¼Ò«¬(nested data models)¡B²V¦X¼Ò«¬(mixed models), ÀH¾÷¼Æ(random coefficient)¡BÀH¾÷®ÄªG¼Ò«¬(random-effects models)¡BÀH¾÷°Ñ¼Æ¼Ò«¬(random parameter models)©Îsplit-plot designs¡C ©w¸q¡G²V¦X®ÄªG ²V¦X®ÄªG¡×©T©w®ÄªG¡ÏÀH¾÷®ÄªG ©T©w®ÄªG(fixed effect)¬O©Ò¦³²Õ¤¤®ÄªG³£¬Û¦P(which are the same in all groups). ÀH¾÷®ÄªG(random effect)¬O¦U²Õ¤§¶¡ªºÀH¾÷§e²{®ÄªG(³£¤£¦P¦P)(which vary across groups). ¦b²V¦X¼Ò«¬(mixed models)¤¤¡A¨CÓlevels³£«Ü©ú½T¦s¦bÀH¾÷©M¨t²Î(©T©w)®ÄªG¡C
2-1-1¦h¼h¦¸¼Ò«¬(¶¥¼h½u©Ê¼Ò«¬HLM)ªº¿³°_ ¦h¼h¦¸¼Ò«¬(multilevel model)¡A¤SºÙ¶¥¼h½u©Ê¼Ò«¬(Hierarchical Linear Modeling, HLM)¡CHLM¦b¥Íª«²Îp»â°ì²ßºDºÙ§@½u©Ê²V¦X¼Ò«¬(Linear Mixed Model, LMM)¡A¦bÀ³¥Î²Îp»â°ì«h±`ºÙ¬°¦h¼h¦¸¼Ò«¬©Î¦h¼h¦¸°jÂk(Multilevel Model / Multilevel Regression)¡A¦ý¤£ºÞ¦p¦óºÙ©I¥¦¡A¨äI«áªºì²z¤jP¬O®t¤£¦hªº¡C ¦h¼h¦¸¼Ò«¬±`¦s¦bªººÃ°Ý¦³¤GÃþ¡G(1)¸ê®Æ¬°¡u¶¥¼h©Ê¡vªº©Ê½è¡C(2)¸ê®Æ¡u«½Æ´ú¶q¡vªº¬ã¨s³]p¡C¦b¥Íª«Âå¾Ç/±Ð¨|µ¥ªÀ·|/¦ÛµM¬ì¾Ç»â°ì¤¤¡A©â¼Ë(smapling)³]p±`±`¦s¦b¡u¶¥¼h©Ê¡v¡A¨Ò¦p¤À¼hÀH¾÷©â¼Ëªk¡A¥¦´N¨Ï¥Î¶¥¼h(Hierarchical)©â¼Ë/¶°¸s©â¼Ë(cluster sampling)¡C¤À¼hÀH¾÷©â¼Ëªk¥i¯à¥H¾Ç®Õ¬°©â¼Ëªº³æ¦ì¡AÀ˵ø«°¶m®t¶Z¹ï¾Ç¥Í¾Ç·~¦¨´Nªº¼vÅT¡A¦¹®É¾Ç¥Í¬O±_ª¬©Î±_ª¬(nested)¦b¾Ç®Õ¤§¤U¡F©Î¬O²Õ´©Î5Mªº¦æ¬°¬ã¨s¤]±`±`¥H¤£¦P¤½¥qªºû¤u¶ñµª°Ý¨÷¸ê®Æ¡A¦¹®Éû¤u¤]¬O±_ª¬¦b¤½¥q¤§¤U¡C¦Ó³o¥H¶Ç²Îªº²Îp¤èªk(¨Ò¦p½Æ°jÂk©ÎANOVA)³B²z³oºØ¶¥¼h©Ê¸ê®Æ·|¦s¦b¤@¨Ç°ÝÃD¡A¶Ç²Îªº°jÂk³Ì«nªº¤@Ó°²©w(assumption)´N¬O»~®t£`¡u¿W¥ß©Ê¡v¡A¥ç§Y¨CÓ¨ü³XªÌªº¨ÌÅܼÆ(µ²ªGÅܼÆ/¨ÌÅܼÆ)¬O¤¬¬Û¿W¥ßªº¡A¦ý¬O¦P¤@¶¡¾Ç®Õªº¾Ç¥Íªº¯S½è²z½×¤WÀ³¸Ó·|¤ñ¸û¬Û¦ü¡A¦Ó¨Ó¦Û¦P¤@¤½¥qªº¤@¸sû¤u¤]À³¸Ó¨ã¦³¤ñ¸û¬Û¦üªº¯S½è¡A¦¹®ÉY¨Ï¥Î¶Ç²Î°jÂk(STaTa«ü¥O¥]¬Areg¡Bheckpoisson¡Bhetregress¡Bintreg¡Bivpoisson¡Bivtobit¡Bnpregress¡Bqreg¡Bsureg¡Btobit¡Btpoisson¡Btruncreg¡Bzip)¡A¥Ñ©ó¥¼¯à¦Ò¶q¡u¸s²Õ¼h¦¸„³ÓÅé¼h¦¸¡vªº½Õ¸`(¤zÂZ¡Amoderator)¡A¾ÉP½u©Ê°jÂk¦¡¥i¯à²£¥Í¿ù»~ªº±À½×®ÄªG¡A²³æ¨Ó»¡§Y¶Ç²Îªº°jÂkµLªk³B²z¡u¤¬¨Ì©Ê¡vªº¸ê®Æ¡C¦¹®É¨Ï¥ÎHLM«h¥i¥H¦Ò¼{¨C¤@ÓÁ`Åé¼h¦¸³æ¦ì(¸ó°ê¡B¾Ç®Õ¡B¤½¥q¡B¾F©~)¤§¤UªºÓÅé¼h¦¸³æ¦ì(¾Ç¥Í¡Bû¤u¡B¦í¤á)¤¬¬°¬Û¨Ìªº¨Æ¹ê¡C ³æ¼h¦¸(«D¦h¼h¦¸)¤§°jÂkªºSTaTa«ü¥O¦p¤U¡G areg¡G§ó®e©öªº¤èªk¨Ó¾AÀ³¨ã¦³³\¦hµêÀÀÅܼƪº¦^Âk¡C arch¡G±aARCH»~®t¤§°jÂk¼Ò«¬¡C arima¡GARIMA¼Ò«¬¡C boxcox¡GBox-Cox°jÂk¼Ò«¬¡C cnsreg¡G¨ü½u©Ê°jÂkconstrained linear regression¡C eivreg¡GÅܼƧt»~®t¤§°jÂkerrors-in-variables regression¡C frontier¡Gstochastic frontier¼Ò«¬¡C gmm¡G¼s¸q°Ê®t¦ôpªkgeneralized method of moments estimation heckman¡GHeckman¿ï¾Ü¼Ò«¬¡C intreg¡G°Ï¶¡°jÂkinterval regression¡C ivregress¡G³æ¤@¤è¦¡¤§¤u¨ãÅܼưjÂksingle-equation instrumental-variables regression¡C ivtobit¡G±a¤º¥ÍÅܼƤ§³]°jÂktobit regression with endogenous variables¡C newey¡G±aNewey-West¼Ð·Ç»~¤§°jÂkregression with Newey-West standard errors nl¡G«D½u©Ê³Ì¤p¥¤è¦ôpªknonlinear least-squares estimation¡C nlsur¡G¤èµ{¦¡ªº«D½u¨t²Îestimation of nonlinear systems of equations¡C qreg¡G¦¨¤À°jÂkquantile(including median) regression¡C reg3¡G¤T¶¥³Ì¤p¥¤èªk¤§°jÂkthree-stage least-squares(3SLS) regression¡C rreg¡G±a±j°·»~®t¤§°jÂka type of robust regression¡C sem¡G µ²ºc¤èµ{¼Ò«¬structural equation models¡C sureg¡G¦ü¤£¬ÛÃö°jÂkseemingly unrelated regression¡C tobit¡G³]°jÂktobit regression¡C(°]ª÷¡B¥ÍÂå¬É±`¥Î¥¦) treatreg¡G³B²z®ÄªG¼Ò¼Ò«¬treatment-effects model¡C truncreg¡GºI§À°jÂktruncated regression¡C xtabond¡GArellano-Bond½u©Ê°ÊºApanel-dataªk(linear dynamic panel-data estimation)¡C xtdpd¡G½u©Ê°ÊºApanel-dataªk(linear dynamic panel-data estimation)¡C xtfrontier¡Gpanel-data stochastic frontier¼Ò«¬¡C xtgls¡Gpanel-data GLS¼Ò«¬¡C xthtaylor¡G»~®t¦¨¤À¤§Hausman-Taylorªk¤Hestimator for error-components models) ¡C xtintreg¡Gpanel-data interval°jÂk¼Ò«¬¡C xtivreg¡Gpanel-data instrumental-variables(2SLS) °jÂk¡C xtpcse¡G±apanel×¥¿¼Ð·Ç»~¤§½u©Ê°jÂk(linear regression with panel-corrected standard errors)¡C xtreg¡G©T©w®ÄªG/ÀH¾÷®ÄªG½u©Ê¼Ò«¬fixed- and random-effects linear models xtregar¡G±aAR(1)¤zÂZ¤§©T©w®ÄªG/ÀH¾÷®ÄªG½u©Ê¼Ò«¬(fixed- and random-effects linear models with an AR(1) disturbance)¡C xttobit¡Gpanel-data³]¼Ò«¬(tobit models)¡C°]ª÷¡B¥ÍÂå¬É±`¥Î¥¦¡C ¦¹¥~¡AMLM¤´¥i¦ôpÁ`Åé¼h¦¸¦h¸ÑÄÀÅܼƹï©óÓÅé¼h¦¸¸ÑÄÀÅܼƪº¼vÅT¡A¨Ò¦p¡A²Õ´»â¾É¬ã¨s¡A¦pªG¦³¤@®a¤½¥qªº²Õ´»â¾É«D±`¨}¦n¡A¥i¯à·|¥¿¦V¼W±jû¤uªº¤u§@ÁZ®Ä¡A¦ý§Ú̬O¦p¦ó´ú¶q¤@®a¤½¥qªº²Õ´»â¾É©O¡H³q±`§ÚÌ·|¥H³o®a¤½¥q©Ò¦³û¤uªº²Õ´»â¾É§@Á`¥´î(group-mean centering)·í¦¨¤½¥q²Õ´»â¾É®Ä¯à¡A¦]¦¹¡A¼vÅTû¤uªº¤u§@ÁZ®Äªº¦]¤l¡A°£¤Fû¤uÓ¤Hªº¤u§@º¡¨¬(ÓÅé¼vÅT)¡A¤]¦³¥i¯à¬O³o®a¤½¥qªº¥§¡²Õ´»â¾É®Ä¯à(Á`Åé¼vÅT)¡C
|
|
|
|