作者查詢 / chien533
作者 chien533 在 PTT [ Statistics ] 看板的留言(推文), 共459則
限定看板：Statistics
看板排序：
全部Gossiping2153Statistics459MLB198NY-Yankees197Baseball122CMWang92JOJO81Blog48Ikariam45sex45HCKuo42Little-Games34joke33PttLifeLaw31AfterPhD29HatePolitics29NewYork23Tokusatsu23japanavgirls21NCAA20NBA19feminine_sex18DIABLO16Steam16movie14Asian-MLB13media-chaos13BBSmovie12PathofExile11PublicIssue11Astros10Boy-Girl9ComeHere9Heroes8PeopleSeries8RedSox8HSNU_10087PCSH91_3057Sangokumusou6ShowOff6TigerBlue6CCF5cksh79th015CPBL5FJU-ACCR945HSNU_8205sandlot_ball5SuperHeroes5DET_Tigers4KOTDFansClub4KS92-3134KS95-3184Monkeys4Padres4specialman4Stephen4Cardinals3CGI-Game3Cobras3CSMU-HSA963CSMU-MED933CSMU-MED953CTSH923013Cubs3CYCUEL95A3ENG_BASE3FCU_EE97A3FJU_JCS113GMAT3GRE3Indians3Jacky_Woo3KingdomHuang3KS94-3023KS94-3203L_TalkandCha3LasVegas3Mariners3MLBGAME3NIU-ECE94b3NKFUST-CCE903NTUST-EE-B953PCCU-CS3Prospect3PttHistory3SFGiants3share3TKU_EE_92C3Translation3WuLing46-3173A-Rod2ADS2AllTogether2AOE2AOSO_Lab2ASHS-93-li2ask2Athletics2Aviation2Baki2BBSview2Blue_Jays2Braves2C_Chat2Cavaliers2ck46th3312ck50th3232CSMU-MED942CTSH943012CYAC91C2DiabloEX2FJU2FJU-Laws942GuardRookies2Hate2HK-movie2HotBloodYuan2HSNU_10552HSNU_10652HSNU_9202IA2ILSH_923122KMT2KNU2KS91-3192KS95-3102KS96-3052KS98-3022maevsoftball2marvel2Mudran2NCCU_SEED2NCHU-FS982NCKU-physiol2NCKU_EARTH982NDMC-P922NHSH_Alumnus2NSwitch2NSYSU-BBTeam2NTHU_STAT942NTOU-MME-99B2NTU-K52NTUCivilism2NTUT_en489b2NTUT_ME495A2NUTN_MS992Patent2Sabermetrics2SCU_Law101B2Soft_Job2StupidClown2TA_AN2TFSHS66th3072THU_BA20002TTU-AMath2USC2VibWang2wegoJT3022Whales2WRADPE2YCSH_alumni2YOLO2Yup02-092Agr-Football1Agronomy-871AIR_JORDAN1Ancient1AndyLau1AppleDaily1Arthropoda1ASIA-uni1basketballTW1bass1Beauty1biker1BLAZERS1BLEACH1Bon_Jovi1BuRuRadio1Canada1car1CCSH_86_3121CCSH_88_3051CCSH_91_3201CCU_EPARC1CFantasy1Cga1CGU_EE981ChineseMed1Christianity1CHSH-93-3041CHU_ME92C1ck51st3211ck56th3241ck57th3081ck57th3181ckkindness1cksh76th191cksh78th151cksh79th3101cksh81st3211cksh82nd3211cksh83rd3091cksh83rd3141cksh83rd3221cksh84th3121cksh84th3231cksh85th1161cksh85th3101cksh85th3171CLHS-53-131CLHS-55-3171CLHS-55-3201CLVSC1CMU_M491CosmosPeople1CourtBasebal1CPU_CID7311CPU_FC7211CPU_FC7611CPULAW1CS15th3441CSI1CSMU-AFL1CSMU-D881CSMU-D891CSMU-D951CSMU-HSA951CSMU-LS891CSMU-MED901CSMU-MIS931CSMU-PT901CSMU-SG1CTSH913011CTSHSub523011CyberFormula1CYCU_CE92B1CYCU_EE_07A1CYCU_MIS_93A1Darvish1digitalk1Dolly1Doraemon1DPP1DTJH_3221DummyHistory1Eagles1EAseries1Elephants1emprisenovel1Emulator1ESOE-971ESP1EZsoft1Facebook1FCU_ECON_93B1FCU_EE97B1FCU_Talk1FinalFantasy1firsttime1FJU-ACC90a1FJU-ACC93a1FJU-EE-2001a1FJU-LAW20101FJU-Laws951FJU-MBA961FJU-Stat95A1FJU-STAT95B1FJU_JCS71FJU_Psy981FSHS-92-3101FXM91-3291gallantry1GameDesign1GIEE_BASKET1Gintama1GOODSTARS1Guardians1GuessX31HCHS593051HCHS603091HCSH1HCSH_10th3121HCSHch13_3111HFU_me_89B1HHHS-961historia1HLHS5thU1HLHS_10thU1home-sale1HS26th3031HSNU_10371HSNU_10581HSNU_10631HSNU_10851HSNU_10951HSNU_11181HSNU_11491HSNU_9891HY29th3071IAM_BioChip1ILSH-943131image1Instant_Mess1Japanese-B951jawawa1JH30th3061JohnnysJr1K_baseball1kawaii1KG95-3041Koei1KS87-3021KS90-3031KS90-3091KS91-3021KS92-3121KS93-3201KS94-3071KS94-3101KS94-3171KS94-3181KS94-3211KS95-2061KS95-3011KS95-3061KS95-3071KS95-3081KS95-3141KS95-3161KS95-3171KS96-2021KS96-3011KS96-3121KS96-3141KS97-2161KS97-3081KS97-3141KS97-3161KS97-3181KS_PMAC1KSHS_Talk1LAND_BASKET1Law-Fala1LawBasketbal1LED_MOUTH1Lei-Tien1Magazine1MartialArt1MartialArts1maruko1MCU-LAW1MDscience6th1MenTalk1Militarylife1MIN-Twins1MKSH-95-61MLB-TW1money1Nanofan1Nantou1Nationals1NBAEasyChat1NCCU04_MAT1NCCU04_Stat1NCCU04_TUR1NCCU05_HISTO1NCCU06_EA1NCCU06_PSYCH1NCCU07_Arab1NCCU07_EA1NCCU07_LAND1NCCU07FMGRAD1NCCU08_LAND1NCHU-CE-421NCHU-FS1001NCHU-MKT991NCHU-photo1NCHU-TIM951NCHU-Wind1NCNUEM1NCTU-ME931NCTU_IIM_BB1NCUT1NCUT-IM1NCYU_DVM_981NDHU-MBA951NDHU-phy951NDHU-phy991NDHU_ACC_BT1NDMC-D621NDMC-D641NDMC-F81NDMC-guitar1NDMC-M1051NDMC-SoftB1NeedFood1NEW_ROC1NFU1NH11th3053091NHLUE-EED9621NHLUE-ILT1NHSH13th3051Nicholas1NTHU-EE-CAPT1NTHU-MSE091NTHU_IEEM-941NTHU_LST_961NTHU_pingpon1NTNU_bridge1ntnuch95c1NTPU-COECO921NTPU-CSIE931NTPU-JLAW941NTPU-LCLS871NTPU-STAT941NTSU-SM951NTTU_CSIE991NTU-Archery1NTU-EM911NTU-EM921NTU-GIIB20041NTU_4H_Club1NTUAC951NTUBSE-B-931NTUCH-901NTUDrama951NTUE-TGD971NTUE_Nse981NTUEE-MBElab1NTUGIEE_EDA1NTUGIPO_PNSL1NTUmed911NTUniNews1NTUST-IM-B941NTUST-TX-B921NTUST-TX-B961NTUT_CH585A1NTUT_IPET4951NUK_AC981NUU_ER1NUU_ME1NY-Mets1Oasis1Oceanography1Odoko-juku1Okajima1Olympics_ISG1Palmar_Drama1Paul_59-1T1pccu_ee891PCGC1PCSH92-3031PCSH97_3131pesoftball1PHAT-MEN1PingTung1PlayStation1Poker1PolBadminton1PoliticBM1PublicBM1PublicServan1PushDoll1Reds1RSSH94_3061ScienceNote1SciMix1SCU-ECG-951SCU_ACC_96A1SCU_ACCM951SCU_chem901SCU_Japan96B1scu_transfer1Seal_Online1SFFamily1SFHS1Shaiya1SHU_LawGrad1Sony-style1Sorry_Youth1SouthPark1SP2_Baseball1SpongeBob1Sportcenter1SSSH-16th-Fk1SuperIdol1SW_Job1TaichungCont1TaiwanDrama1Taoyuan1TCFSHvolley1Teacher1Tech_Job1Tennis1TFSHS64th3091TFSHS64th3181TFSHS65th3161TFSHS65th3201TFSHS66th3191TFSHS67th3051TFSHS68th3161TFSHS68th3211TFSHS69th3121TFSHS69th3181TheMatrix1THU-CHE931THU-P-Softbo1THU_PH901THUID_Grad1TKU-IC941TKU-IE941TKU_BF93C1TKU_HisSB1TKU_IHM1TNFSH87111transgender1Tsao1TSH93_11TSSHS62th3181TTSH-11th3021TTU-AFL1TTU-I93A1TTU-ME941TW-history1TW_Entertain1TWproducts1tyart1TYSH46-3021TYSH48-3011TYSH49-3021UTAH-JAZZ1V_ScHooL1VET_921Web_Design1Weird_Sound1Wenhua1WhiteSox1Wikipedia1WorldCup1Wrestle1XiangSheng1Yanzi1YCChen1YoungDotx31YP92-3121YP96-3141Yunlin1YZU_MBA941<< 收起看板(549)
16F推: 人數就是count data, 所以當然是Poisson分配。09/01 07:00
15F推: 統計只是工具，重點是你論文題目的創新度。題目很鳥但09/01 06:57
16F→: 用了個逆天無敵的統計方法也沒啥鳥用。09/01 06:57
11F推: Q1:比較有可能是complete separation或quasi-complete07/19 02:51
12F→: separation，共線性的特徵大多是估計量反轉（正負相反）07/19 02:52
13F→: Q2: 回答同Q107/19 02:52
14F→: Q3:ＣＩ會很大表示點估計不穩定，結果就不穩健07/19 02:53
15F→: Q4:羅吉斯迴歸沒有自己檢定共線性的方法，比較可行的就07/19 02:54
16F→: 是把原始資料用一般線性迴歸來跑，然後拿他的VIF值來用07/19 02:55
9F推: 統計裡面adjust的意思是應把變數放入模型裡面，怎麼會是06/18 11:26
10F→: 拿掉？我是覺得你老師根本不懂這個字的統計意義，只是06/18 11:26
11F→: 單純賣弄英文，所以你還是請你老師講中文吧！06/18 11:26
3F推: 妳得先講清楚不同質的定義是指什麼06/18 11:21
4F→: 如果是指不同單位，那相加起來當然不妥06/18 11:22
5F→: 妳試試看先做標準化後再相加起來再比較吧！06/18 11:23
20F推: 你這應該把醫生設定成隨機變項，讓每個醫生跟所有醫生06/18 10:57
21F→: 的平均來比才有意義。設定dummy variable得預設某個醫06/18 10:57
22F→: 生當作reference level, 但你準備要讓哪個醫生當referen06/18 10:57
23F→: ce level? 亦或是哪個醫生夠格當reference level?這是你06/18 10:57
24F→: 得在文章內回答的問題。06/18 10:57
3F推: 你這明顯是模型沒收斂，所以出現一個近乎於無限寬的信06/18 10:51
4F→: 賴區間，所以顯示這個信賴區間完全沒有意義。你的模型得06/18 10:51
5F→: 重做。06/18 10:51
8F推: 你這是時間序列資料，一般的邏輯斯迴歸處理不了時間相06/18 10:44
9F→: 關性，得用generalized linear mixed model06/18 10:44
10F推: 或者是generalized additive model(若想用time smoother06/18 10:45
11F→: )06/18 10:45
2F推: 北大蠻著重在生統? Not really....12/20 03:46
3F→: They focus on finance.12/20 03:47
10F推: mixed model正解07/16 03:37