{"id":103,"date":"2020-12-20T09:34:40","date_gmt":"2020-12-20T09:34:40","guid":{"rendered":"https:\/\/datasciencebasic.com\/?p=103"},"modified":"2021-03-14T09:37:24","modified_gmt":"2021-03-14T09:37:24","slug":"mot-so-khai-niem-xac-suat-thong-ke-co-ban-thuong-su-dung-trong-phan-tich-du-lieu-phan-2","status":"publish","type":"post","link":"https:\/\/datasciencebasic.com\/?p=103","title":{"rendered":"M\u1ed9t s\u1ed1 kh\u00e1i ni\u1ec7m x\u00e1c su\u1ea5t th\u1ed1ng k\u00ea c\u01a1 b\u1ea3n th\u01b0\u1eddng s\u1eed d\u1ee5ng trong ph\u00e2n t\u00edch d\u1eef li\u1ec7u (Ph\u1ea7n 2)"},"content":{"rendered":"\n

Ti\u1ebfp theo ph\u1ea7n 1, trong b\u00e0i vi\u1ebft n\u00e0y, t\u00f4i s\u1ebd kh\u00e1i qu\u00e1t th\u00eam m\u1ed9t s\u1ed1 kh\u00e1i ni\u1ec7m x\u00e1c su\u1ea5t th\u1ed1ng k\u00ea c\u01a1 b\u1ea3n th\u01b0\u1eddng s\u1eed d\u1ee5ng trong ph\u00e2n t\u00edch d\u1eef li\u1ec7u.<\/p>\n\n\n\n

3.<\/strong> Quan h\u1ec7 gi\u1eefa c\u00e1c bi\u1ebfn<\/strong><\/p>\n\n\n\n

C\u00e1c th\u01b0\u1edbc \u0111o th\u1ed1ng k\u00ea d\u1ef1a tr\u00ean m\u1ed1i quan h\u1ec7 gi\u1eefa c\u00e1c bi\u1ebfn l\u00e0 T\u01b0\u01a1ng quan (correlation) v\u00e0 Hi\u1ec7p ph\u01b0\u01a1ng sai (covariance). M\u1ed9t s\u1ed1 ng\u01b0\u1eddi th\u01b0\u1eddng nh\u1ea7m l\u1eabn t\u01b0\u01a1ng quan v\u1edbi quan h\u1ec7 nguy\u00ean nh\u00e2n \u2013 k\u1ebft qu\u1ea3. L\u00e0 m\u1ed9t nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u, b\u1ea1n c\u1ea7n ph\u00e2n bi\u1ec7t \u0111\u01b0\u1ee3c nh\u1eefng \u0111i\u1ec3m kh\u00e1c bi\u1ec7t gi\u1eefa hai kh\u00e1i ni\u1ec7m n\u00e0y.<\/p>\n\n\n\n

T\u01b0\u01a1ng quan (correlation)<\/em><\/strong><\/p>\n\n\n\n

T\u01b0\u01a1ng quan gi\u1eefa hai bi\u1ebfn th\u1ec3 hi\u1ec7n m\u1ed1i quan h\u1ec7 gi\u1eefa ch\u00fang, r\u1eb1ng khi m\u1ed9t bi\u1ebfn t\u0103ng l\u00ean th\u00ec n\u00f3 s\u1ebd t\u00e1c \u0111\u1ed9ng nh\u01b0 th\u1ebf n\u00e0o \u0111\u1ebfn bi\u1ebfn kia. Khi ch\u00fang ta n\u00f3i hai bi\u1ebfn c\u00f3 t\u01b0\u01a1ng quan thu\u1eadn v\u1edbi nhau, \u0111i\u1ec1u \u0111\u00f3 c\u00f3 ngh\u0129a l\u00e0 n\u1ebfu m\u1ed9t bi\u1ebfn t\u0103ng th\u00ec bi\u1ebfn kia c\u0169ng t\u0103ng theo. Khi gi\u00e1 tr\u1ecb t\u01b0\u01a1ng quan gi\u1eefa hai bi\u1ebfn g\u1ea7n b\u1eb1ng 0 c\u00f3 ngh\u0129a l\u00e0 kh\u00f4ng c\u00f3 nhi\u1ec1u m\u1ed1i quan h\u1ec7 gi\u1eefa ch\u00fang. D\u01b0\u1edbi \u0111\u00e2y l\u00e0 m\u1ed9t s\u1ed1 v\u00ed d\u1ee5 v\u1ec1 bi\u1ec3u \u0111\u1ed3 scatter plot v\u1edbi c\u00e1c m\u1ee9c \u0111\u1ed9 t\u01b0\u01a1ng quan kh\u00e1c nhau c\u1ee7a 2 bi\u1ebfn<\/p>\n\n\n\n

\"\"<\/figure><\/div>\n\n\n\n

Ngu\u1ed3n: https:\/\/www.mathsisfun.com\/data\/correlation.html<\/a><\/p>\n\n\n\n

Quan h\u1ec7 nguy\u00ean nh\u00e2n \u2013 k\u1ebft qu\u1ea3 (causation)<\/em><\/strong><\/p>\n\n\n\n

Nh\u01b0 \u0111\u00e3 \u0111\u1ec1 c\u1eadp \u1edf tr\u00ean, nhi\u1ec1u ng\u01b0\u1eddi nh\u1ea7m l\u1eabn gi\u1eefa t\u01b0\u01a1ng quan v\u00e0 quan h\u1ec7 nh\u00e2n qu\u1ea3, ngh\u0129a l\u00e0 khi c\u00e1c bi\u1ebfn \u2018A\u2019 v\u00e0 \u2018B\u2019 c\u00f3 t\u01b0\u01a1ng quan cao, h\u1ecd cho r\u1eb1ng A khi\u1ebfn B x\u1ea3y ra, nh\u01b0ng c\u00f3 th\u1ec3 kh\u00f4ng ph\u1ea3i nh\u01b0 v\u1eady. Tr\u00ean th\u1ef1c t\u1ebf, s\u1ef1 t\u01b0\u01a1ng quan kh\u00f4ng gi\u1ea3i th\u00edch b\u1ea5t c\u1ee9 \u0111i\u1ec1u g\u00ec li\u00ean quan \u0111\u1ebfn nh\u00e2n qu\u1ea3. Ch\u1eb3ng h\u1ea1n, khi nhi\u1ec7t \u0111\u1ed9 kh\u00f4ng kh\u00ed t\u0103ng, s\u1ed1 ng\u01b0\u1eddi ph\u1ea3i nh\u1eadp vi\u1ec7n t\u0103ng v\u00e0 doanh s\u1ed1 b\u00e1n kem c\u0169ng t\u0103ng l\u00ean. V\u00ec v\u1eady, \u1edf \u0111\u00e2y doanh s\u1ed1 b\u00e1n kem v\u00e0 s\u1ed1 ng\u01b0\u1eddi ph\u1ea3i nh\u1eadp vi\u1ec7n c\u00f3 m\u1ed1i quan h\u1ec7 t\u01b0\u01a1ng quan thu\u1eadn. Nh\u01b0ng doanh s\u1ed1 b\u00e1n kem kh\u00f4ng tr\u1ef1c ti\u1ebfp g\u00f3p ph\u1ea7n l\u00e0m gia t\u0103ng s\u1ed1 ng\u01b0\u1eddi ph\u1ea3i nh\u1eadp vi\u1ec7n, ho\u1eb7c ng\u01b0\u1ee3c l\u1ea1i. \u1ede \u0111\u00e2y, nhi\u1ec7t \u0111\u1ed9 kh\u00f4ng kh\u00ed t\u0103ng m\u1edbi l\u00e0 nguy\u00ean nh\u00e2n c\u1ee7a c\u1ea3 hai hi\u1ec7n t\u01b0\u1ee3ng tr\u00ean.<\/p>\n\n\n\n

V\u00ec v\u1eady, \u0111\u1eebng bao gi\u1edd \u0111\u00e1nh gi\u00e1 hai bi\u1ebfn c\u00f3 quan h\u1ec7 nh\u00e2n qu\u1ea3 ch\u1ec9 v\u00ec ch\u00fang c\u00f3 s\u1ef1 t\u01b0\u01a1ng quan v\u1edbi nhau.<\/p>\n\n\n\n

4. Ph\u00e2n ph\u1ed1i (distribution)<\/strong><\/p>\n\n\n\n

L\u00e0 m\u1ed9t nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u, b\u1ea1n c\u1ea7n bi\u1ebft c\u00e1c ph\u00e2n ph\u1ed1i (distribution) kh\u00e1c nhau v\u00ec n\u00f3 s\u1ebd gi\u00fap hi\u1ec3u r\u00f5 h\u01a1n v\u1ec1 t\u1eadp d\u1eef li\u1ec7u v\u00e0 c\u00f3 th\u1ec3 ch\u1ecdn m\u1ed9t m\u00f4 h\u00ecnh ph\u00e2n t\u00edch d\u1ef1 b\u00e1o th\u00edch h\u1ee3p. C\u00f3 nhi\u1ec1u d\u1ea1ng ph\u00e2n ph\u1ed1i c\u1ee7a d\u1eef li\u1ec7u, nh\u01b0ng \u00edt nh\u1ea5t b\u1ea1n c\u1ea7n bi\u1ebft c\u00e1c d\u1ea1ng ph\u00e2n ph\u1ed1i sau \u0111\u00e2y:<\/p>\n\n\n\n

\"\"<\/figure><\/div>\n\n\n\n

Ngu\u1ed3n: https:\/\/365datascience.com\/tutorials\/statistics-tutorials\/different-distribution\/<\/a><\/p>\n\n\n\n

Ph\u00e2n ph\u1ed1i chu\u1ea9n (normal distribution)<\/em><\/strong> – Ph\u00e2n ph\u1ed1i chu\u1ea9n c\u00f3 ngh\u0129a l\u00e0 h\u1ea7u h\u1ebft c\u00e1c quan s\u00e1t t\u1eadp trung g\u1ea7n gi\u00e1 tr\u1ecb trung b\u00ecnh v\u00e0 khi ch\u00fang ta di chuy\u1ec3n ra xa gi\u00e1 tr\u1ecb trung b\u00ecnh, s\u1ed1 l\u01b0\u1ee3ng quan s\u00e1t gi\u1ea3m \u0111i v\u00e0 ph\u00e2n ph\u1ed1i l\u00e0 \u0111\u1ed1i x\u1ee9ng, ngh\u0129a l\u00e0 ph\u1ea7n b\u00ean tr\u00e1i v\u00e0 b\u00ean ph\u1ea3i c\u1ee7a gi\u00e1 tr\u1ecb trung b\u00ecnh l\u00e0 t\u01b0\u01a1ng t\u1ef1 nhau.<\/p>\n\n\n\n

Ph\u00e2n ph\u1ed1i \u0111\u1ed3ng \u0111\u1ec1u (uniform distribution)<\/em><\/strong> – Trong tr\u01b0\u1eddng h\u1ee3p ph\u00e2n ph\u1ed1i \u0111\u1ed3ng \u0111\u1ec1u, x\u00e1c su\u1ea5t x\u1ea3y ra l\u00e0 nh\u01b0 nhau \u0111\u1ed1i v\u1edbi t\u1ea5t c\u1ea3 c\u00e1c l\u1ef1a ch\u1ecdn. Ch\u1eb3ng h\u1ea1n nh\u01b0 khi ta gieo 1 con x\u00fac x\u1eafc, x\u00e1c su\u1ea5t ta c\u00f3 m\u1ed7i gi\u00e1 tr\u1ecb t\u1eeb 1 \u0111\u1ebfn 6 l\u00e0 nh\u01b0 nhau.<\/p>\n\n\n\n

Ph\u00e2n ph\u1ed1i nh\u1ecb th\u1ee9c (binomial distribution)<\/em><\/strong> – L\u00e0 m\u1ed9t ph\u00e2n ph\u1ed1i t\u1ea7n su\u1ea5t c\u1ee7a c\u00e1c k\u1ebft qu\u1ea3 c\u00f3 th\u1ec3 c\u00f3 t\u1eeb m\u1ed9t t\u1eadp h\u1ee3p h\u1eefu h\u1ea1n. V\u00ed d\u1ee5 khi ta gieo hai con x\u00fac x\u1eafc, c\u00e1c gi\u00e1 tr\u1ecb c\u00f3 th\u1ec3 n\u1eb1m trong kho\u1ea3ng t\u1eeb 2 \u0111\u1ebfn 12 v\u00e0 m\u1ed7i gi\u00e1 tr\u1ecb n\u00e0y \u0111\u1ec1u c\u00f3 m\u1ed9t gi\u00e1 tr\u1ecb x\u00e1c su\u1ea5t x\u1ea3y ra. Bi\u1ec3u di\u1ec5n c\u00e1c gi\u00e1 tr\u1ecb x\u00e1c xu\u1ea5t n\u00e0y l\u00ean bi\u1ec3u \u0111\u1ed3 histogram, ta s\u1ebd th\u1ea5y \u0111\u01b0\u1ee3c h\u00ecnh \u1ea3nh c\u1ee7a m\u1ed9t ph\u00e2n ph\u1ed1i nh\u1ecb th\u1ee9c.<\/p>\n\n\n\n

5. \u0110\u1ecbnh l\u00fd gi\u1edbi h\u1ea1n trung t\u00e2m (Central Limit Theorem)<\/strong><\/p>\n\n\n\n

\u0110\u1ecbnh l\u00fd Gi\u1edbi h\u1ea1n Trung t\u00e2m l\u00e0 m\u1ed9t kh\u00e1i ni\u1ec7m ph\u1ed5 bi\u1ebfn trong th\u1ed1ng k\u00ea, theo \u0111\u1ecbnh l\u00fd n\u00e0y, khi ta l\u1ea5y c\u00e0ng nhi\u1ec1u m\u1eabu t\u1eeb m\u1ed9t qu\u1ea7n th\u1ebf, ph\u00e2n ph\u1ed1i m\u1eabu s\u1ebd c\u00e0ng g\u1ea7n v\u1edbi ph\u00e2n ph\u1ed1i chu\u1ea9n (normal distribution), v\u1edbi gi\u1ea3 \u0111\u1ecbnh r\u1eb1ng c\u00e1c m\u1eabu \u0111\u1ec1u c\u00f3 k\u00edch th\u01b0\u1edbc gi\u1ed1ng nhau.<\/p>\n\n\n\n

N\u00f3i m\u1ed9t c\u00e1ch kh\u00e1c, theo \u0111\u1ecbnh l\u00fd gi\u1edbi h\u1ea1n trung t\u00e2m, v\u1edbi k\u00edch th\u01b0\u1edbc m\u1eabu \u0111\u1ee7 l\u1edbn t\u1eeb m\u1ed9t t\u1ed5ng th\u1ec3 c\u00f3 m\u1ee9c ph\u01b0\u01a1ng sai (variance) h\u1eefu h\u1ea1n, gi\u00e1 tr\u1ecb trung b\u00ecnh c\u1ee7a t\u1ea5t c\u1ea3 c\u00e1c m\u1eabu t\u1eeb c\u00f9ng m\u1ed9t t\u1ed5ng th\u1ec3 s\u1ebd x\u1ea5p x\u1ec9 b\u1eb1ng gi\u00e1 tr\u1ecb trung b\u00ecnh c\u1ee7a t\u1ed5ng th\u1ec3. Ngo\u00e0i ra, t\u1ea5t c\u1ea3 c\u00e1c m\u1eabu s\u1ebd tu\u00e2n theo m\u1ed9t ph\u00e2n ph\u1ed1i g\u1ea7n v\u1edbi ph\u00e2n ph\u1ed1i chu\u1ea9n, v\u1edbi c\u00e1c ph\u01b0\u01a1ng sai x\u1ea5p x\u1ec9 b\u1eb1ng ph\u01b0\u01a1ng sai c\u1ee7a t\u1ed5ng th\u1ec3 chia cho k\u00edch th\u01b0\u1edbc c\u1ee7a m\u1ed7i m\u1eabu. Trong th\u1ef1c t\u1ebf, k\u00edch th\u01b0\u1edbc m\u1eabu b\u1eb1ng ho\u1eb7c l\u1edbn h\u01a1n 30 \u0111\u01b0\u1ee3c coi l\u00e0 \u0111\u1ee7 l\u1edbn \u0111\u1ec3 tu\u00e2n theo \u0111\u1ecbnh l\u00fd gi\u1edbi h\u1ea1n trung t\u00e2m, c\u00f3 ngh\u0129a l\u00e0 ph\u00e2n ph\u1ed1i c\u1ee7a m\u1eabu \u0111\u01b0\u1ee3c coi l\u00e0 g\u1ea7n v\u1edbi ph\u00e2n ph\u1ed1i chu\u1ea9n.<\/p>\n","protected":false},"excerpt":{"rendered":"

Ti\u1ebfp theo ph\u1ea7n 1, trong b\u00e0i vi\u1ebft n\u00e0y, t\u00f4i s\u1ebd kh\u00e1i qu\u00e1t th\u00eam m\u1ed9t s\u1ed1 kh\u00e1i ni\u1ec7m x\u00e1c su\u1ea5t th\u1ed1ng […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"nf_dc_page":"","om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[3],"tags":[],"class_list":["post-103","post","type-post","status-publish","format-standard","hentry","category-kien-thuc"],"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"","_links":{"self":[{"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/posts\/103","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=103"}],"version-history":[{"count":2,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/posts\/103\/revisions"}],"predecessor-version":[{"id":108,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/posts\/103\/revisions\/108"}],"wp:attachment":[{"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=103"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=103"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=103"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}