20.500.12556/RUL-186152 Vzročnost v analizi preživetja Causal inference in survival analysis Analiza preživetja je statistična metoda za proučevanje časa do pojava dogodka. V praksi so podatki, ki jih želimo analizirati z analizo preživetja, pogosto opazovalni. To pomeni, da je ocenjevanje vzročnega učinka obravnavane terapije ali izpostavljenosti oteženo zaradi prisotnosti motečih spremenljivk. Klasične metode analize preživetja, kot sta metodi Kaplan–Meier in Coxov model sorazmernih tveganj, praviloma ne omogočajo ustreznega vzročnega sklepanja v takšnih okoliščinah. To predstavlja pomemben izziv, ki je spodbudil razvoj metod vzročnega sklepanja, prilagojenih za analizo preživetja. V teoretičnem uvodu smo predstavili temeljne koncepte vzročnega sklepanja, med drugim Simpsonov obrat, načine prikazovanja odvisnosti med spremenljivkami, glavne oznake ter ključne predpostavke vzročne analize. Sledil je pregled teorije analize preživetja, kjer smo opisali osnovne količine ter klasični metodi Kaplan–Meier in Coxov model sorazmernih tveganj. V zadnjem delu teoretičnega poglavja smo obravnavali metode vzročnega sklepanja, ki se uporabljajo v analizi preživetja: G-formulo, IPTW in AIPTW. Te metode omogočajo upoštevanje motečih spremenljivk in ocenjevanje vzročnih učinkov v opazovalnih podatkih. V empiričnem delu smo izvedli obsežen sklop simulacij, v katerih smo primerjali omenjene metode s klasično neparametrično cenilko Kaplan–Meier. Najprej smo ponovili simulacijo iz izbrane literature, nato pa zasnovali dodatne scenarije, kjer smo postopoma spreminjali strukturo generiranja podatkov, vpliv spremenljivk, porazdelitev terapije ter pravilnost modeliranja mehanizma izida in terapije. Rezultati so pokazali, da metode, ki ustrezno upoštevajo moteče dejavnike, vodijo do manj pristranskih ocen kot klasični pristopi. G-formula se izkaže kot občutljiva na napačno modeliranje izida, IPTW na napačno modeliranje terapije, metoda AIPTW pa deluje dobro, ko je pravilno specificiran vsaj eden izmed prej navedenih modelov. Ko pa sta oba modela nepravilno specificirana, se njena zanesljivost zmanjša. Dodatne simulacije so pokazale, da neupoštevanje relevantnih spremenljivk (kršitev predpostavke o izmerjenih motečih dejavnikih) vodi do pristranskosti vseh obravnavanih metod, pri čemer so razlike med njimi relativno majhne. Prav tako se je izkazalo, da so metode, ki temeljijo na uteževanju (IPTW in AIPTW), bolj občutljive na neravnovesje v deležih terapij, kar se odraža v večji variabilnosti ocen, zlasti pri redkih terapijah. Rezultati magistrskega dela prispevajo k boljšemu razumevanju uporabe in primerjave metod vzročnega sklepanja v analizi preživetja ter nakazujejo, da premišljena in skrbna uporaba teh metod lahko pomembno izboljša kakovost sklepov, pridobljenih iz opazovalnih podatkov. Survival analysis is a statistical framework for studying the time to the occurrence of an event. In practice, data subjected to survival analysis are often observational, which implies that the causal effect of a treatment or exposure of interest cannot be directly inferred due to the presence of confounding variables. Classical survival analysis methods, such as the Kaplan–Meier estimator and the Cox proportional hazards model, generally do not support valid causal inference in such settings. This limitation represents a significant challenge and has motivated the development of causal inference methods specifically adapted to survival analysis. In the theoretical introduction, we presented the fundamental concepts of causal inference, including Simpson’s paradox, representations of dependencies between variables, standard notation, and key assumptions underlying causal analysis. This was followed by a review of survival analysis theory, in which we described the basic quantities of interest as well as the classical Kaplan–Meier estimator and the Cox proportional hazards model. The final part of the theoretical chapter focused on causal inference methods applied in survival analysis, namely the G-formula, inverse probability of treatment weighting (IPTW), and augmented inverse probability of treatment weighting (AIPTW). These methods enable adjustment for confounding variables and facilitate the estimation of causal effects in observational data. In the empirical part, we conducted an extensive simulation study in which we compared the aforementioned methods with the classical nonparametric Kaplan–Meier estimator. First, we replicated a simulation study from the selected literature and subsequently designed additional scenarios in which we systematically varied the data-generating process, the strength of covariate effects, the treatment assignment distribution, and the correctness of model specification for both the outcome and treatment mechanisms. The results demonstrated that methods which appropriately account for confounding yield less biased estimates than classical approaches. The G-formula was shown to be sensitive to misspecification of the outcome model, and IPTW to misspecification of the treatment model, whereas AIPTW performed well provided that at least one of the two models was correctly specified. When both models were misspecified, however, its performance deteriorated. Further simulations indicated that failure to adjust for relevant covariates (i.e., violation of the assumption of no unmeasured confounding) leads to biased estimates across all considered methods, with relatively small differences between them. In addition, weighting-based methods (IPTW and AIPTW) were found to be more sensitive to imbalance in treatment proportions, resulting in increased variability of the estimates, particularly in settings with rare treatments. The results of this master’s thesis contribute to a deeper understanding of the application and comparison of causal inference methods in survival analysis and suggest that careful and judicious use of these methods can substantially improve the quality of conclusions drawn from observational data. vzročnost analiza preživetja G-formula IPTW AIPTW Causal inference Survival Analysis G-formula IPTW AIPTW true false false Slovenski jezik Angleški jezik Magistrsko delo/naloga 2026-08-27 15:00:19 2026-08-27 15:00:35 2026-08-28 04:41:59 0000-00-00 00:00:00 2026 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 1970-01-01 63026 Brlogar_Petra_-_Vzrocnost_v_analizi_prezivetja.pdf Brlogar_Petra_-_Vzrocnost_v_analizi_prezivetja.pdf 1 B0EFFDDF38A73A7A8C245D1CA697386F 36810fcd3a8dd68167e5869cd233fa80261e34e0d7893e9f7ea774777aef913c e1b347c2-a216-11f1-9b0d-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=245262 Fakulteta za elektrotehniko 0 0 0