writ No.2 | Explain andor W's’ 4 icftnuaors - _ vis Wetheds - z eae “pns- A rsndom:= no: 8s a yo! chosen by chance. Rom a Sek “OF possible Aiatues - be ae >|Tk Nos no paren % each no: has an equal probaly og being, Selecled- Be Simulation 4S 4 _jechwique used fo wnodel xea)~world _ simulatfons: Using ‘Ydndom no"s - Bet ce TH help us understand how stm ight Work _urithout eS actualy, Performing \he ‘Yeal process - Se at ia Methods - of _Stinslarfon i .. ___ 9] Monte carlo Simvlaifon= pare ;—=Uses Ydndom sampling 5. Staitefes: 40 model _ ———+ ___| Complex problems 2 | Ee : = eq: Predicring _Mne Stock Markel i. A __ © Ofscroie —_ Crent _ Simutaifon = ____ | _=Simutores events at Spedife mes: ley. Modeling A queue’ tna hawk ox hospital - 9) Stochaske _Simu\dtion 7 2 = Tpwowes Xandomness — ty Mne Sim: eg: Weasner _predicfon models _ Dererminseae Xrnuraiton =o randomness Snvoived- ~ Same ie enti always give. Ye Same OP: 4 ——_jeq. calcularing Avavel Mme based on fedl__speed & Nssance- _ ee alnae a pe SAS = | = Binds Mnes best soi? among many possibte on 6 ee with _uNcertoinay ~Shows ——o renior bos in gy —@) wt Code Evetnation) | = etnsss valttdatforn 7 “ander data splits RM iasigl Bcacng sc ©] Simuasion_of real —Wer'd system S Simul to oaaae Weather ox —queuctng systems;an | ° ° 1 - RD Bedi hypothesis esting with Ws Aypess 7” | fns-_ eieesio Aesiing Ie a “stasieal wnethod! Used to _ check {Fa latte” ox ascumelfon abeut-a_ population 3s] tue or not usr, Sample data e } There ake shoe 6! yo\me: tr _Lhdt!- ‘ F Nan_hypothests (ie) - pated >The deta claim - No change ox ‘io. effect — | Alternative Wyporhedts (a) Fo ~MWere Is a_change oy effect cao eet lunes: e @ Z-test = Used when Sample Size %e large popl?_so- 4s : known: a =n3l : 6 checking aNerdge Wt: of Trems tna: large Shipment __| ® F-test = used when sample size fe Small ox _popl? $2 %s_| unknown: Suypes © one-Sample © 4w0-Sdmpie _© paired. © | chi-square test Oe" test) ‘ ~used “fox categorical data ___|- checks heather Obsewed data fis expected dala - eg. Is \nere a ‘veiatvonship: bet? gendex & Noting preference F @ Propostton test =Tesis whediher a’ popl proportion Ch) Sc, equal to. La specific Values €q. checking it 60%» Of people _prefer ted _cver coffee: - Bf visualized using - : © bar chats Ss ® box PissNisualted “using box plots e bos. PSE Olicts-cquneutests ai -sele tno : — | _=Tesis_wheather there's a relationship bet? 9 caegrted | Naatahles or tt chsewed data Rio.expecied dota: » sya eelnamesaa Oso. for _Geodnesa— of - ale > “o:8: yo O= ~obsenied.- Req - 4 i € = Expected freq. eq. Does. gender cattect? cH product choice 2 ni ep Vables, box chans etc: : mpaxe: Yartantes. ce 00 Lore eee At fo sel “hen thecting i€ 2 sarmples | | Simkida_voriabi | -ofen used fr ANOVA: Pe voronme-t Now: of gxp- 2Rea Updates probability based om ‘net ——— “use 3. sGnes! Meorern - : 5 |. as gives the probabsitey ora saints being, Ane ne | Sipe AAejeaing or accep king he i oo ea PLOWS PCW) eo) Explain Yemecsion wiih WS. types: eos. TR Onte: | | Lee Smite ‘aye ‘on al b P Ca\o) = Fy iP OY Pong S | Used fy _ predictive Braden aL mented! atagn ise |NiSualired with Probabiliag Atonbutons , posierfox graphs: __Ans- Regression“ fs_a_stasitcal method used to Xhe— | _Aeldvonship bet? a dependent Vartabie § One tox Tnaxe_ independent Variables: TW. help us oe. Ox estinate a value based on_ known ddta: = eg. You want $0 _predfck shudents tanks braved —on = hours studied + GS \inea Regression i predics lp using a Staafght ae Dis rmok ©. y 5: Predicinng Salort based on iid oF rae re ©] Multiple — Knoar Regression | ~ uses more cae one Up Vattable. | egh= Y= Ot Gad -Gn%a by tse Predicting house — — afrenalnnninh! & An. ne op a porns fio) Série egmee shertiys aie tiene WS a ugh Be pees 2. Gauksaih process: i i + | Reames: