Ranch, Hees Modelt - is a stahshval mefbo usec 40 model the relatenship behveen a cleperdent variable and one or moe independent viable. The main aim is 7 understanel how chanyes in the iodependent variables oe associakel with chetges ia oleh voclent Variable ' key components of vyression ye ' Dependent Variable : thisisthe variable Put nreds 4 be prebcteel This also knoun as respnge vuriable 2-Indeperdent Varables: Theo variables hue aa intleene on the dependent variables. Tris also known as rechedw variables 2 -Rapesion Eyeboo! repens tHe rlabuatbip befveen uo varably ” fiw simple linear regression with ene indegendert varake S.. ¥= dependent yeoia ble ma 7 ne She Beg X =Todlependnt teroble , ie o Ope Po Jolewept fy wake chen Xs eI 9] 4 Residvals:. Th is hee ditlewcis é behuren — th ce ° ie obsewed salves of ss a we onl the wales preclerol 4 He epasion & Ase mp hens. # passion Analysis relies on ascuen phen depend BB ap oes ph nc Ho alas Venstort voriome of eres and ny Slnaiy, F 5 Model Evalvhen’ Quression vovllels oe evalvakel usm various wre hus suchas R* adjukel R>, Meon syrarecl Erov J % FT pes of Regression lie Reluborshia hehueen varabls is asemod tebe LINEAR iAulkple - Sare as Iieear but mulhple independent rarabes imohect, eter hehuren varabes in wow LIMeAR Mlogehe -7 : This is usecl wher the depenckod verable is Bivapy dud it predicts the Prbably of eaacae of un evend‘ “XN as, FileHay, oh Resse Eq sachin: J well the node) Fits He dah, Tr vimehes, assessiog how andl how yliable V the eshmates of the ‘oethiicnds ene The wmrcn methods are * 1. Cethcient Fstmale: eae the eshte coethcients Cslepes and iolercept) te undlerstond the divechien encl wrupritule of the relaboaships between the —inclepenclent ancl cle pendent varia bls 2 hee - of - Bt measures '. * R?Ctoethaent of Detemmoaben) i Th repeserk the prperheo | of vareme in the dependent — variuble thet is explamect | by the independent varrables , Higher R? vales inclines betby Ft A) Adjusted 2% Similar to R% bub it pomalres fy the rember of predicts 0 the model, which ton be vseh) uben ey models with litlerent nembes of predhichns > Residual Analasts | A) Residval Plots: Plot He residvals (the cliflevene behuren the | obsered and prechicledl vale) against The — prechctel vale oy aqviast each indepenlent eriable . ook for pucHerns oy herds | Athe —yesiduals that right indicate vivlufens of regression | assemphins (eg heterostedosh«ity ) non heeeri ty ) ») Narolily of Residtals's Assess whether He residials are approximately namally dishbuded This can be chechect oe Dates 9 @ plots ov Shchsheo) Fests 4 Hedero sce dosheity). Chek whether the vurene of the wichals | is constant ee diHerent levels of the rdepenceat | variables, This con be assessecl visrall ahee> vesicle) plots f qGP nee 4. Ihcollineari ty’ se tela linee? ¢ Wee conelahens — Whueen independent voniables he deo fy melbiollamrity, which ours wher inde prodent vanubls me highly meld with each other. Tt tan jaf latecl sey) enors ond rouke loefhuent — eshnuts coreliable 6 predictive perfomance! Assess how well the — rode! prachicts new date b usie Htbngues Suh os Cvs vahduhen ow splitheg te dala ino Peicicg ond cates sek. 7 Signitioe ies Condut hy potess tests w delermine whether the coctbuerts ave t ritiantly cliflerent fare 200. The rast remmen desk is t-fest vs iodnidval Coethuents, whith assess wheter Cech —prechetw bus SigetBeond etel on depentlent rarable |8 Odhiners and Led hental Porodss. | DB may hue a dispwpaberatly large impact co the | eee Tek hese we inthenbal interns of they levenge w ther eHect on the repre ion cook ha erk Steps te fe beilk —veqression mode! Collech ® Date erie | atter clude om winables 40 eesti de [Date pepeesiq. - Cleen Re clade, honclle miss; rales PXrpliatey Dade Avelyas: Unelestead the velabeship, behueen vanabs anh siummarin oe Model Selahen'- choose a ppwprale | Nahwe of date l#) Model rn, SPIE He larly into twinig gael deshap sets MAodel Lvaladen Frulude the pertrorcae of he § d, 4) Ace Deployment. One subhed with the models perkrmeoe , dep tf prule pPreclichens on new dak Coe model basecl on the mode t1" | Correlahon ag ig concerned with Ba ees toy the Shength ve the velabenshio — befween variables s \ ‘orrlahon model: ae dn this, both X and ¥ variables are yandem vanables The sample observahens are obkuoe! by selechay a voadem sample of the units of associahon and Za keg on Cath a meusurement of X anc a meascrement one, ovthae to build = cone ben mode! Date preparahun' Orgam ze pe cedu 50 that, och row veprsen Parr of varikbks and incheles velevont Potwes Hud ae influene the corelahien —ehuren Hose vanables 3) Fechoe Fevineerin IT den’ tee featves thet ge be prediche of the wrlahen behveen variables. “These sacldes the properhes of the vanables, dlishrbuhens ,veluhenghips el #) Model Selecher’ choose an appmprate eee model fer Dredeteg the tonelaton sales This woudl be a hoor ihe model ov move complex rnodlels . -dpodig a ne ioe ot the clata #) Mode) Traaming Split the dd i; into tele ard feshoy sets MAode) Ervaladion: Geneon sysarteck enor) \ balate the perforane vey MSE, RMSE ov R? delerminahen 2) Mode Toterprelation! Analyze the wethuents tp cackeistoncl which Pahves ae mnt preliihe of the —forrelation behueen variables [Medel Dephy ment 1 Depby it tw precicd coelaheng fer pews pars of varia bles ‘Correlation Cockhecient : Tt 5 a Stubsta) measue thet quohhes te Shength and direchon of te velabonship — behuten hue vanables Jt 15 dlenotel by the symbol yr and venge$ d f 4 foro -1 toy Covelaben toethaent ¥, = g 2 ems ya [Em ely) vie an. ave n ten Coothivsent Ie hich covelaben (OF x ond Y vara bles for w é oe where, Kandy are the ymmans of X and ¥ cor) Covela ton. cothuent Y, = Cov Op Sx S Where, d COvarione , erlny) z 2xy ae = xg n nen, foie VEG! Gisd stendorch ee mee ae [2 -@ aIndications of — fonelation toePhvient . | Y21, perkct pesihie tonelakin . As one wrrable totrenses, the other vanable also increuses lineos < 4 re-1, perfect negare forelahin | As one yanable increases the otter vaviuble decreases sey » 250 6 Mo tonelakon) > There i, np linear velahurshy W-\ | Y=) % OLN oo A Y=O. / Pere peste posite re vacmekidel date y 4 beact ative nega tive Shes covets bon Pr Ree cali aneWee Ree ean 7 Ripon , Qrelaton for detu vsog R/ Pifthen. pylon offer powerful Vibes for pectooing R oJ Yeoression — analysis \ Y USiog kk N deed Data: H# oad data <- read. tsv ("dat csv’ Perfrra — Regression: DAS Ake + Perdam — |incor reqressivo XQ rnedel Z- Im (dependent_ variable ~ independent _vorablei + independent - varale >, clara = cata) Summon Croce) d Calulke Coreen Leet forebton — mahix £- Cor (data) Prot Coorelatin —rrats ). | doak Date: ae A import Pandas os pa A dood — dutaset Creplce “duo. csv” with your File path ) date = pd. reed csv (“dafa.esv")Pe rfovm Req yession + import stutsrnodels.api as oro 4 Defre independet and dependent vana bles Xe dade [Ciodeperdent vasable’, odependent varia ble a 4 # Add consent term to iedepenclnt varrables X= sm.add — const Cx) (# BE he vapesum model mode) = oe CY x) ALO | Prot Seemes oF the Een yeasts priot Cenodlel. SEG aD) Caludate Correlahen # (alulite coveluton reatox Covrelahin_ rnukix = clack (ore) prot (tonelation maker)