Multistate models STK4080 H Competing risk setting 2. Illness-Death setting 3. General event histories and Aalen-Johansen estimator

Size: px
Start display at page:

Download "Multistate models STK4080 H Competing risk setting 2. Illness-Death setting 3. General event histories and Aalen-Johansen estimator"

Transcription

1 Multistate models p. 1/36 Multistate models STK4080 H16 1. Competing risk setting 2. Illness-Death setting 3. General event histories and Aalen-Johansen estimator

2 Multistate models p. 2/36 Multistate models Will consider stochastic processes X(t) that can move between states{0,1,...,k} with 1 transition ratesα ij (t) = lim h 0 P(X(t+h) = j X(t) = i) h under a Markov assumption, F s is the history up to time s, P(X(t+h) = j X(s) = i) = P(X(t+h) = j F s ) We will then derive the Aalen-Johansen estimator of P ij (s,t) = P(X(t) = j X(s) = i). We will consider The competing risk setting (model) The illness-death (e.g. healthy-illness-death)) model The general case

3 Competing risks Denotes the state "alive" by 0 and state "death from causeh" by h(= 1,2,...,k) Transition (hazard) rate α 0h (t) of cause h k causes of death (can only observe one cause) Will callα 0h (t) cause-specific hazard rates Multistate models p. 3/36

4 Multistate models p. 4/36 Competing risks, random variables Postulates T h = time until death of causeh = 1,2,...,k. Assumes that the T h are independent with hazardsα 0h (t). Note: Observes only T = min(t 1,T 2,...,T k ) and D = h if T = T h, not the differentt h s. However, the framework is suitable for describing the model for the process X(t) on states {0, 1,..., k} with transition rates α 0h (t) from the "alive" state 0 to cause of death states h = 1,...,k. In particular we find that P 00 (s,t) = P(X(t) = 0 X(s) = 0) = P(T > t T > s) = exp( k h=1 α s 0h(u)du)

5 Multistate models p. 5/36 Cumulative incidence functions The transition probabilities for competing risks are given as P 0h (s,t) = P(X(t) = h X(s) = 0) = s P 00 (s,u)α 0h (u)du with the reasoning that to be in state h attone have stayed in state 0 from time s until some time u where0 s < u < t and then moved to state h at time u. Afteruthe process has to stay in h. We refer to P 0h (s,t) as cumulative incidence functions. Necessarily P 00 (s,t)+p 01 (s,t)+ +P 0k (s,t) = 1.

6 Kolomogorov equations for P 0h (s,t) The P 0h (s,t) can be alternatively derived from the Kolmogorov (forward) equations t P 00(s,t) = P 00 (s,t) k α 0h (t) h=1 and t P 0h(s,t) = P 00 (s,t)α 0h (t) The first equations leads to P 00 (s,t) = exp( k h=1 s α 0h (u)du) and integrating up the second we get P 0h (s,t) = P(X(t) = h X(s) = 0) = s P 00 (s,u)α 0h (u)du Multistate models p. 6/36

7 Multistate models p. 7/36 Competing risks, censoring The process X(t) is observed up to some censoring time C. Observation is summarized by T = min(t,c), where T is the event time of X(t) without censoring, and D = h if X( T) = h. In particular D = 0 if T = C. With n independent individual processesx i (t) we observe ( T i,d i ). We get the following counting process framework: Y 0 (t) = #{ T i t)} = no. still at risk N 0h (t) = #{ T i t,d i = h)} = counts deaths of causeh N(t) = k h=1 N 0h(t) total no. of deaths before t

8 Multistate models p. 8/36 Intensity processes and martingales Assume independent censoring. Then we have the following intensity processes for the counting processes: Intensity process ofn 0h (t) becomesλ 0h (t) = Y 0 (t)α 0h (t) Intensity process ofn 0 (t) becomes λ 0 (t) = Y 0 (t) k h=1 α 0h(t) = Y 0 (t)α 0 (t) We then obtain martingales M 0h (t) = N 0h (t) Λ 0h (t) = N 0h (t) 0 Y 0(s)α 0h (s) M 0 (t) = N 0 (t) Λ 0 (t) = N 0 (t) 0 Y 0(s)α 0 (s)ds In particular the M 0h (t) are orthogonal (uncorrelated) because N 0 (t) only jumps with step 1, implying that the N 0h (t) will not jump at the same time t.

9 Multistate models p. 9/36 Cumulative hazardsa 0h (t) = 0 α 0h(s)ds can be estimated on competing risk data by Nelson-Aalen estimators Â0h(t) = dn 0h (s). 0 Y 0 (s) We then just treat events of type j h as censoring. We get a martingale representation, J(s) = I(Y 0 (s) > 0), Â 0h (t) = 0 J(s)α 0h (s)ds+ so the last term has expectation zero and 0 dm 0h (s) Y 0 (s) Var(Â0h(t)) Var(Â0h(t) A 0h(t)) = E 0 α 0h (s)ds Y 0 (s) The Â0h(t) indicates the shape of the α 0h (t) and by kernel smoothing one can obtain estimated cause-specific hazards.

10 Multistate models p. 10/36 "Survival functions" S 0h (t) = exp( A 0h (t))? We could estimate S 0h (t) by exp( Â0h(t) or by a type Kaplan-Meier estimator Ŝ0h(t) = s t (1 dn 0h(s) Y 0 (s) ). Under the specification of the competing risk model through independent T h with hazardα 0h (t) these estimate the survival function oft h and have the interpretation as the survival function if all other causes are eradicated. However: We can only observe the minimum of the T h and it is not possible to tell if event times will be independent. Thus Ŝ 0h (t) should be interpreted with extreme care. The1 Ŝ0h(t) do NOT estimate cumulative incidences P 0h (0,t)!

11 Multistate models p. 11/36 Estimation of survival function P 00 (s,t) Since N 0 (t) has intensity process Y 0 (t)α 0 (t) we get that an almost unbiased estimator of the survival function P 00 (s,t) = exp( α s 0(u)du) is given by P 00 (s,t) = [ 1 dn ] 0(u), Y 0 (u) s<u t corresponding directly to the Kaplan-Meier estimator. The properties: expectation, variance and asymptotical normality, follows exactly in the same way as for Kaplan-Meier. In particular ˆP 00 (s,t) has expectation, (J(s) = I(Y 0 (s) > 0)), E[exp( s J(u)α 0 (u)du)] = E[P 00(s,t)].

12 thus close to unbiased. Multistate models p. 12/36 Estimated cumulative incidence function We noted P 0h (s,t) = s P 00 (u)α 0h (s,u)du, thus by plug-in estimates of the cumulative incidence functions are given by ˆP 0h (s,t) = s ˆP 00 (s,u ) dn 0h(u) Y 0 (u). The representation N 0h (t) = Y 0 0(s)α 0h (s)ds+m 0h (t) leads to ˆP 0h (s,t)] = s ˆP 00 (s,u )J(u)α(u)du+ s where the last term is a martingale and so ˆP 00 (s,u ) dm 0h(u) Y 0 (u) E[ˆP 0h (s,t)] = s E[P 00(s,u )J(u)]α 0h (u)du,

13 Multistate models p. 13/36 Variances of ˆP 0h (s,t) are more tricky This is because in the representation ˆP 0h (s,t) = s ˆP 00 (s,u )J(u)α 0h (u)du+ s ˆP 00 (s,u ) dm 0h(u) Y 0 (u) there is (essential) randomness in both terms, that is also in s ˆP 00 (s,u )J(u)α 0h (u)du. This was not the case for the Nelson-Aalen estimator: Var(Â0h(t)) Var(Â0h(t) A 0h(t)) = E 0 α 0h (s)ds Y 0 (s) However, a variance formula for the cumulative incidence function ˆP 0h (s,t) is given in ABG, eq. (3.89).

14 Multistate models p. 14/36 Simulation competing risk n = 1000 individuals Cause 1 hazardα 01 (t) = t, Weibull with k = 2 and b = 1 Cause 2 hazardα 02 (t) = t 0.5, Weib. k = 0.5 and b = 1 Censoring C U[0,1] Calculate ˆP 00 (t) by Kaplan-Meier ignoring cause Y 0 (t) and dn 0h (t) from n.risk and n.event in cause-specific survfit Cumulative incidence function by cumsum (I should have fixed this slightly, I use ˆP 00 (s) instead of ˆP 00 (s ))

15 par(mfrow=c(1,2)) plot(survfit1,fun="event",mark.time=f,conf.int=f,ylim=c(0,1)) lines(stepfun(survfit1$time,cuminc1),lty=2) title("cause 1") plot(survfit2,fun="event",mark.time=f,conf.int=f,ylim=c(0,1)) lines(stepfun(survfit1$time,cuminc2),lty=2) legend(0.1,0.1,c("1-km","cum.incidence"),lty=1:2,bty="n") Multistate models p. 15/36 Simulation competing risk in R n<-1000 time1<-rweibull(n,2) time2<-rweibull(n,0.5) censtime<-runif(n)*2 obstime<-pmin(time1,time2,censtime) d<-1*(obstime==time1)+2*(obstime==time2) survfit1<-survfit(surv(obstime,d==1) 1) survfit2<-survfit(surv(obstime,d==2) 1) survfit0<-survfit(surv(obstime,d>0) 1) cuminc1<-cumsum(survfit0$surv*survfit1$n.event/survfit1$n.risk) cuminc2<-cumsum(survfit0$surv*survfit2$n.event/survfit2$n.risk)

16 Multistate models p. 16/36 Simulation competing risks: ˆP 0h (t) and 1 Ŝ0h(t) Clearly 1-Kaplan-Meier does not estimate cumulative incidence! Cause 1 Cause KM cum.incidence

17 Multistate models p. 17/36 Stepfun from Splus stepfun<-function(datax, datay, type = "left") { # augment a set of points so that it plots as a # left-continuous function # allow both (x,y) and (structure with $x $y) input if(missing(datay)) x <- datax$x else x <- datax if(missing(datay)) y <- datax$y else y <- datay n <- length(x) type <- charmatch(type, c("left", "right")) if(is.na(type)) stop("the type must be left or right continuous") if(any(diff(x) < 0)) stop("the x vector must be sorted") if(type == 2) { x <- rev(x) y <- rev(y) }

18 Multistate models p. 18/36 Stepfun from Splus, second half if(n > 2) { # remove redundant points dupy <- c(t, diff(y[ - n])!= 0, T) dupx <- c(t, diff(x[ - n])!= 0, T) x <- x[dupx & dupy] y <- y[dupx & dupy] n <- length(x) } #create the step function xrep <- rep(x[2:n], rep(2, n - 1)) yrep <- rep(y[1:(n - 1)], rep(2, n - 1)) if(type == 1) list(x = c(x[1], xrep), y = c(yrep, y[n])) else list(x = c(rev(xrep), x[1]), y = c(y[n], rev(yrep))) }

19 Simulation competing risks: Cumulative hazards survfit1<-survfit(surv(obstime,d==1) 1,type="fh2") survfit2<-survfit(surv(obstime,d==2) 1,type="fh2") plot(survfit1,fun="cumhaz",mark.time=f,main="cum.haz. 1") plot(survfit2,fun="cumhaz",mark.time=f,main="cum.haz. 2") Cum.haz. 1 Cum.haz We see thatα 01 (t) is increasing while α 02 (t) is decreasing. Multistate models p. 19/36

20 Multistate models p. 20/36 Estimation of cumulative incidence in R Previously one would need the R library mstate to estimate cumulative incidence. Now it has become possible to use the standard R library survival by adding the option type="mstate" to the survfit command. > estcuminc=survfit(surv(obstime,d,type="mstate") 1) > names(estcuminc) [1] "n" "time" "n.risk" "n.event" [5] "n.censor" "prev" "p0" "transitio [9] "cumhaz" "std.err" "istate" "lower" [13] "upper" "conf.type" "conf.int" "states" [17] "type" "call" > plot(estcuminc,conf.int=t,ylim=c(0,0.62))

21 Multistate models p. 21/36 Cumulative incidence plots using survfit Plots of ˆP 0h (t) with confidence intervals

22 Multistate models p. 22/36 (Healthy-)Illness-Death (ID) Hazards (intensities) α gh (t) for transition from state g to state h at time t with α 01 (t) > 0, α 02 (t) > 0 and α 12 (t) > 0 (at least for some t) and all other α gh (t) = 0.

23 Transition probabilities in ID-process LetX(t) be the state of the process at time t,x(t) {0,1,2} and define transition probabilities We then get that and P gh (s,t) = P(X(t) = h X(s) = g) P 00 (s,t) = exp( P 11 (s,t) = exp( P 01 (s,t) = s s (α 01 (u)+α 02 (u))du), s α 12 (u)du) P 00 (s,u)α 01 (u)p 11 (u,t)du where the integrand is the "density" of staying in 0 until u, then moving to 1 atuand staying in 1 up to t. Multistate models p. 23/36

24 Multistate models p. 24/36 Transition probabilities in ID-process, II We may also calculate P 02 (s,t) = s P 00 (s,u)α 02 (u)du+ s P 01 (s,u)α 12 (u)du where the first terms come from the direct move from 0 to 2 (death without previous disease) and the second from death after disease. Finally P 12 (s,t) = s exp( u s α 12(v)dv)α 12 (u)du = 1 exp( s α 12(u)du) = 1 P 11 (s,t) and all other P gh (s,t) = 0 (except forp 22 (s,t) = 1).

25 Multistate models p. 25/36 Kolmogorov forward eq. for ID Then P t 00(s,t) = P 00 (s,t)(α 01 (s)+α 02 (s)) P t 01(s,t) = P 00 (s,t)α 01 (s) P 01 (s,t)α 12 (s) P t 02(s,t) = P 00 (s,t)α 02 (s)+p 01 (s,t)α 12 (s) P t 11(s,t) = P 11 (s,t)α 12 (s) The first equation gives P 00 (s,t) = exp( s (α 01(u)+α 02 (u))du) The second is a 1.order inhomogeneous differential equation The third is just integrating up functions already derived The fourth clearly give P 11 (s,t) = exp( s α 12(u)du)

26 Estimation transition probabilities, ID Let Y h (t) = no. in state h at time t N gh (t) = no. of direct transitions fromg to h in [0,t] N g (t) = h N gh(t) no. transitions out of g in [0,t] We get estimates and ˆP 00 (s,t) = s<u t [1 dn 0 (u) ] Y 0 (u) ˆP 11 (s,t) = s<u t [1 dn 12(u) ] Y 1 (u) ˆP 01 (s,t) = ˆP s 00 (s,u) dn 01(u) Y 0 ˆP (u) 11 (u,t) ˆP 02 (s,t) = s ˆP 00 (s,u) dn 02(u) Y 0 (u) + s ˆP 01 (s,u) dn 12(u) Y 1 (u) Multistate models p. 26/36

27 General (complicated) event scheme Multistate models p. 27/36

28 Multistate models p. 28/36 General event schemes Many states May move back and forth between states, for instance Healthy Disease Married Unmarried Out of workforce In workforce LetX(t) be the state of the process att α gh (t) = hazard/intensity of moving fromg to h att P gh (s,t) = P(X(t) = h X(s) = g) = transition probabilities and the matrix of transition probabilities P(s,t) = [P gh (s,t)] k g,h=0

29 General event schemes, contd. With this setup we have a Markov process, thus for a partition, 0 = t 0 < t 1 < < t K we have P(t 0,t K ) = P(t 0,t 1 ) P(t 1,t 2 ) P(t K 1,t K ) Furthermore P gh (u,u+du) = α gh (u)du when g h and P gg (u,u+du) = 1 α gh (u)du = 1 α g (u)du h g thus with α(u) given as the matrix with α g (u) along the diagonal and α gh (u) outside the diagonal we may write P(u,u+du) = I+α(u)du Multistate models p. 29/36

30 Multistate models p. 30/36 General event schemes, estimation Thus the matrix of transition probabilities may be written as a continuous product P(s,t) = s<u t P(u,u+du) = s<u t [I+α(u)du] and estimated by using just the identity matrix I for I+α(u)du at timesuwith no events and dâ(u) are dâgh(u) Here I+dÂ(u) where the elements of = dn gh(u) Y g (u) for g h dâgg(u) = dn g (u) Y g (u) Y h (t) = no. in state h at time t N gh (t) = no. of direct transitions fromg to h in [0,t] N g (t) = h N gh(t) no. transitions out of g in [0,t]

31 Multistate models p. 31/36 General event schemes, estimation The estimator of the transition probabilities ˆP(s,t) = [I+dÂ(u)] s<u t is called the Aalen-Johansen estimator Note that without ties in the data the matrix I+dÂ(u) will be diagonal except for a row g where the diagonal element equals 1 1/Y g (u) and one other element equals 1/Y g (u) (see ABG, eq. 3.79).

32 Multistate models p. 32/36 Aalen-Johansen estimator, simulation The Aalen-Johansen estimator may be applied to the Healthy-Illness-Death process. Let α 01 (t) = t (Weibull, k = 2) α 02 (t) = t (Weibull, k = 2) α 12 (t) = 1 (exponential) Censoring uniform on [0,1] n = 1000 individuals

33 Multistate models p. 33/36 Aalen-Johansen estimator, R-code, simulation ID n<-100 timesick<-rweibull(n,2) timedeath<-rweibull(n,2) timesickdeath<-rexp(n) censtime<-runif(n)*2 D01<-1*(timesick<pmin(timedeath,censtime)) D02<-1*(timedeath<pmin(timesick,censtime)) D12<-1*(D01==1)*((timesick+timesickdeath)<censtime) D1C<-1*(D01==1)*(censtime<(timesick+timesickdeath)) obstimes<-c(timesick[d01==1],timedeath[d02==1], (timesick+timesickdeath)[d12==1]) events<-c(rep(1,sum(d01)),rep(2,sum(d02)),rep(3,sum(d12))) events<-events[order(obstimes) ] obstimes<-sort(obstimes)

34 Multistate models p. 34/36 Aalen-Johansen estimator, R-code, counting processes Y0<-numeric(0) Y1<-numeric(0) N01<-numeric(0) N02<-numeric(0) N12<-numeric(0) N1C<-numeric(0) for (i in 1:length(obstimes)){ Y0[i]<-sum(pmin(timesick,timedeath,censtime)>=obstimes[i]) N01[i]<-sum(D01*(timesick<=obstimes[i])) N02[i]<-sum(D02*(timedeath<=obstimes[i])) N12[i]<-sum(D12*((timesick+timesickdeath)<=obstimes[i])) N1C[i]<-sum(D1C*(censtime<=obstimes[i])) } Y1<-0 for (i in 1:(length(obstimes)-1)) Y1<-c(Y1,N01[i]-N12[i]-N1C[i]) dn01<-n01-c(0,n01[1:(length(obstimes)-1)]) dn02<-n02-c(0,n02[1:(length(obstimes)-1)]) dn12<-n12-c(0,n12[1:(length(obstimes)-1)])

35 par(mfrow=c(1,2)) plot(obstimes,pmat[1,1,],type="l",xlab="time",ylab="p0h(t)",ylim=c(0,1)) lines(obstimes,pmat[1,2,],lty=2) lines(obstimes,pmat[1,3,],lty=4) legend(0.5,1,c("h=0","h=1","h=2"),lty=c(1,2,4),bty="n") plot(obstimes,pmat[2,2,],type="l",xlab="time",ylab="p1h(t)",ylim=c(0,1)) lines(obstimes,pmat[2,3,],lty=2) legend(0.5,1,c("h=1","h=2"),lty=c(1,2),bty="n") Multistate models p. 35/36 Aalen-Johansen estimator, R-code, estimator Pmat<-array(dim=c(3,3,length(obstimes))) Qmat<-array(dim=c(3,3,length(obstimes))) for (i in 1:length(obstimes)) { if (Y0[i]>0) line1<-c(1-(dn01[i]+dn02[i])/y0[i],dn01[i]/y0[i],dn02[i]/y0 if (Y0[i]==0) line1<-c(1,0,0) if (Y1[i]>0) line2<-c(0,1-dn12[i]/y1[i],dn12[i]/y1[i]) if (Y1[i]==0) line2<-c(0,1,0) line3<-c(0,0,1) Qmat[,,i]<-t(matrix(c(line1,line2,line3),nrow=3)) if (i==1) Pmat[,,1]<-Qmat[,,1] if (i>1) Pmat[,,i]<-Pmat[,,i-1]%*%Qmat[,,i] }

36 time time Multistate models p. 36/36 Simulation Aalen-Johansen, ID Plots of ˆP gh (t): P0h(t) h=0 h=1 h=2 P1h(t) h=1 h=

Chapter 4 Fall Notations: t 1 < t 2 < < t D, D unique death times. d j = # deaths at t j = n. Y j = # at risk /alive at t j = n

Chapter 4 Fall Notations: t 1 < t 2 < < t D, D unique death times. d j = # deaths at t j = n. Y j = # at risk /alive at t j = n Bios 323: Applied Survival Analysis Qingxia (Cindy) Chen Chapter 4 Fall 2012 4.2 Estimators of the survival and cumulative hazard functions for RC data Suppose X is a continuous random failure time with

More information

Statistics 262: Intermediate Biostatistics Non-parametric Survival Analysis

Statistics 262: Intermediate Biostatistics Non-parametric Survival Analysis Statistics 262: Intermediate Biostatistics Non-parametric Survival Analysis Jonathan Taylor & Kristin Cobb Statistics 262: Intermediate Biostatistics p.1/?? Overview of today s class Kaplan-Meier Curve

More information

Multi-state models: prediction

Multi-state models: prediction Department of Medical Statistics and Bioinformatics Leiden University Medical Center Course on advanced survival analysis, Copenhagen Outline Prediction Theory Aalen-Johansen Computational aspects Applications

More information

Estimating transition probabilities for the illness-death model The Aalen-Johansen estimator under violation of the Markov assumption Torunn Heggland

Estimating transition probabilities for the illness-death model The Aalen-Johansen estimator under violation of the Markov assumption Torunn Heggland Estimating transition probabilities for the illness-death model The Aalen-Johansen estimator under violation of the Markov assumption Torunn Heggland Master s Thesis for the degree Modelling and Data Analysis

More information

Understanding product integration. A talk about teaching survival analysis.

Understanding product integration. A talk about teaching survival analysis. Understanding product integration. A talk about teaching survival analysis. Jan Beyersmann, Arthur Allignol, Martin Schumacher. Freiburg, Germany DFG Research Unit FOR 534 jan@fdm.uni-freiburg.de It is

More information

PhD course in Advanced survival analysis. One-sample tests. Properties. Idea: (ABGK, sect. V.1.1) Counting process N(t)

PhD course in Advanced survival analysis. One-sample tests. Properties. Idea: (ABGK, sect. V.1.1) Counting process N(t) PhD course in Advanced survival analysis. (ABGK, sect. V.1.1) One-sample tests. Counting process N(t) Non-parametric hypothesis tests. Parametric models. Intensity process λ(t) = α(t)y (t) satisfying Aalen

More information

Multistate Modeling and Applications

Multistate Modeling and Applications Multistate Modeling and Applications Yang Yang Department of Statistics University of Michigan, Ann Arbor IBM Research Graduate Student Workshop: Statistics for a Smarter Planet Yang Yang (UM, Ann Arbor)

More information

Exercises. (a) Prove that m(t) =

Exercises. (a) Prove that m(t) = Exercises 1. Lack of memory. Verify that the exponential distribution has the lack of memory property, that is, if T is exponentially distributed with parameter λ > then so is T t given that T > t for

More information

STAT331. Combining Martingales, Stochastic Integrals, and Applications to Logrank Test & Cox s Model

STAT331. Combining Martingales, Stochastic Integrals, and Applications to Logrank Test & Cox s Model STAT331 Combining Martingales, Stochastic Integrals, and Applications to Logrank Test & Cox s Model Because of Theorem 2.5.1 in Fleming and Harrington, see Unit 11: For counting process martingales with

More information

Asymptotic Distributions for the Nelson-Aalen and Kaplan-Meier estimators and for test statistics.

Asymptotic Distributions for the Nelson-Aalen and Kaplan-Meier estimators and for test statistics. Asymptotic Distributions for the Nelson-Aalen and Kaplan-Meier estimators and for test statistics. Dragi Anevski Mathematical Sciences und University November 25, 21 1 Asymptotic distributions for statistical

More information

Multi-state Models: An Overview

Multi-state Models: An Overview Multi-state Models: An Overview Andrew Titman Lancaster University 14 April 2016 Overview Introduction to multi-state modelling Examples of applications Continuously observed processes Intermittently observed

More information

Survival Analysis: Weeks 2-3. Lu Tian and Richard Olshen Stanford University

Survival Analysis: Weeks 2-3. Lu Tian and Richard Olshen Stanford University Survival Analysis: Weeks 2-3 Lu Tian and Richard Olshen Stanford University 2 Kaplan-Meier(KM) Estimator Nonparametric estimation of the survival function S(t) = pr(t > t) The nonparametric estimation

More information

DAGStat Event History Analysis.

DAGStat Event History Analysis. DAGStat 2016 Event History Analysis Robin.Henderson@ncl.ac.uk 1 / 75 Schedule 9.00 Introduction 10.30 Break 11.00 Regression Models, Frailty and Multivariate Survival 12.30 Lunch 13.30 Time-Variation and

More information

Linear rank statistics

Linear rank statistics Linear rank statistics Comparison of two groups. Consider the failure time T ij of j-th subject in the i-th group for i = 1 or ; the first group is often called control, and the second treatment. Let n

More information

STAT Sample Problem: General Asymptotic Results

STAT Sample Problem: General Asymptotic Results STAT331 1-Sample Problem: General Asymptotic Results In this unit we will consider the 1-sample problem and prove the consistency and asymptotic normality of the Nelson-Aalen estimator of the cumulative

More information

Stochastic Modelling Unit 1: Markov chain models

Stochastic Modelling Unit 1: Markov chain models Stochastic Modelling Unit 1: Markov chain models Russell Gerrard and Douglas Wright Cass Business School, City University, London June 2004 Contents of Unit 1 1 Stochastic Processes 2 Markov Chains 3 Poisson

More information

Robust estimates of state occupancy and transition probabilities for Non-Markov multi-state models

Robust estimates of state occupancy and transition probabilities for Non-Markov multi-state models Robust estimates of state occupancy and transition probabilities for Non-Markov multi-state models 26 March 2014 Overview Continuously observed data Three-state illness-death General robust estimator Interval

More information

Nonparametric Model Construction

Nonparametric Model Construction Nonparametric Model Construction Chapters 4 and 12 Stat 477 - Loss Models Chapters 4 and 12 (Stat 477) Nonparametric Model Construction Brian Hartman - BYU 1 / 28 Types of data Types of data For non-life

More information

Estimation for Modified Data

Estimation for Modified Data Definition. Estimation for Modified Data 1. Empirical distribution for complete individual data (section 11.) An observation X is truncated from below ( left truncated) at d if when it is at or below d

More information

Cox s proportional hazards model and Cox s partial likelihood

Cox s proportional hazards model and Cox s partial likelihood Cox s proportional hazards model and Cox s partial likelihood Rasmus Waagepetersen October 12, 2018 1 / 27 Non-parametric vs. parametric Suppose we want to estimate unknown function, e.g. survival function.

More information

Survival Analysis Math 434 Fall 2011

Survival Analysis Math 434 Fall 2011 Survival Analysis Math 434 Fall 2011 Part IV: Chap. 8,9.2,9.3,11: Semiparametric Proportional Hazards Regression Jimin Ding Math Dept. www.math.wustl.edu/ jmding/math434/fall09/index.html Basic Model Setup

More information

Empirical Processes & Survival Analysis. The Functional Delta Method

Empirical Processes & Survival Analysis. The Functional Delta Method STAT/BMI 741 University of Wisconsin-Madison Empirical Processes & Survival Analysis Lecture 3 The Functional Delta Method Lu Mao lmao@biostat.wisc.edu 3-1 Objectives By the end of this lecture, you will

More information

Kernel density estimation in R

Kernel density estimation in R Kernel density estimation in R Kernel density estimation can be done in R using the density() function in R. The default is a Guassian kernel, but others are possible also. It uses it s own algorithm to

More information

Lecture 6 PREDICTING SURVIVAL UNDER THE PH MODEL

Lecture 6 PREDICTING SURVIVAL UNDER THE PH MODEL Lecture 6 PREDICTING SURVIVAL UNDER THE PH MODEL The Cox PH model: λ(t Z) = λ 0 (t) exp(β Z). How do we estimate the survival probability, S z (t) = S(t Z) = P (T > t Z), for an individual with covariates

More information

4 Testing Hypotheses. 4.1 Tests in the regression setting. 4.2 Non-parametric testing of survival between groups

4 Testing Hypotheses. 4.1 Tests in the regression setting. 4.2 Non-parametric testing of survival between groups 4 Testing Hypotheses The next lectures will look at tests, some in an actuarial setting, and in the last subsection we will also consider tests applied to graduation 4 Tests in the regression setting )

More information

STAT 331. Martingale Central Limit Theorem and Related Results

STAT 331. Martingale Central Limit Theorem and Related Results STAT 331 Martingale Central Limit Theorem and Related Results In this unit we discuss a version of the martingale central limit theorem, which states that under certain conditions, a sum of orthogonal

More information

Survival Analysis: Counting Process and Martingale. Lu Tian and Richard Olshen Stanford University

Survival Analysis: Counting Process and Martingale. Lu Tian and Richard Olshen Stanford University Survival Analysis: Counting Process and Martingale Lu Tian and Richard Olshen Stanford University 1 Lebesgue-Stieltjes Integrals G( ) is a right-continuous step function having jumps at x 1, x 2,.. b f(x)dg(x)

More information

A multi-state model for the prognosis of non-mild acute pancreatitis

A multi-state model for the prognosis of non-mild acute pancreatitis A multi-state model for the prognosis of non-mild acute pancreatitis Lore Zumeta Olaskoaga 1, Felix Zubia Olaskoaga 2, Guadalupe Gómez Melis 1 1 Universitat Politècnica de Catalunya 2 Intensive Care Unit,

More information

( t) Cox regression part 2. Outline: Recapitulation. Estimation of cumulative hazards and survival probabilites. Ørnulf Borgan

( t) Cox regression part 2. Outline: Recapitulation. Estimation of cumulative hazards and survival probabilites. Ørnulf Borgan Outline: Cox regression part 2 Ørnulf Borgan Department of Mathematics University of Oslo Recapitulation Estimation of cumulative hazards and survival probabilites Assumptions for Cox regression and check

More information

Definitions and examples Simple estimation and testing Regression models Goodness of fit for the Cox model. Recap of Part 1. Per Kragh Andersen

Definitions and examples Simple estimation and testing Regression models Goodness of fit for the Cox model. Recap of Part 1. Per Kragh Andersen Recap of Part 1 Per Kragh Andersen Section of Biostatistics, University of Copenhagen DSBS Course Survival Analysis in Clinical Trials January 2018 1 / 65 Overview Definitions and examples Simple estimation

More information

Survival analysis in R

Survival analysis in R Survival analysis in R Niels Richard Hansen This note describes a few elementary aspects of practical analysis of survival data in R. For further information we refer to the book Introductory Statistics

More information

STAT331 Lebesgue-Stieltjes Integrals, Martingales, Counting Processes

STAT331 Lebesgue-Stieltjes Integrals, Martingales, Counting Processes STAT331 Lebesgue-Stieltjes Integrals, Martingales, Counting Processes This section introduces Lebesgue-Stieltjes integrals, and defines two important stochastic processes: a martingale process and a counting

More information

Statistical Inference and Methods

Statistical Inference and Methods Department of Mathematics Imperial College London d.stephens@imperial.ac.uk http://stats.ma.ic.ac.uk/ das01/ 31st January 2006 Part VI Session 6: Filtering and Time to Event Data Session 6: Filtering and

More information

Chapter 7: Hypothesis testing

Chapter 7: Hypothesis testing Chapter 7: Hypothesis testing Hypothesis testing is typically done based on the cumulative hazard function. Here we ll use the Nelson-Aalen estimate of the cumulative hazard. The survival function is used

More information

Survival Analysis. Stat 526. April 13, 2018

Survival Analysis. Stat 526. April 13, 2018 Survival Analysis Stat 526 April 13, 2018 1 Functions of Survival Time Let T be the survival time for a subject Then P [T < 0] = 0 and T is a continuous random variable The Survival function is defined

More information

Practice Exam 1. (A) (B) (C) (D) (E) You are given the following data on loss sizes:

Practice Exam 1. (A) (B) (C) (D) (E) You are given the following data on loss sizes: Practice Exam 1 1. Losses for an insurance coverage have the following cumulative distribution function: F(0) = 0 F(1,000) = 0.2 F(5,000) = 0.4 F(10,000) = 0.9 F(100,000) = 1 with linear interpolation

More information

Notes largely based on Statistical Methods for Reliability Data by W.Q. Meeker and L. A. Escobar, Wiley, 1998 and on their class notes.

Notes largely based on Statistical Methods for Reliability Data by W.Q. Meeker and L. A. Escobar, Wiley, 1998 and on their class notes. Unit 2: Models, Censoring, and Likelihood for Failure-Time Data Notes largely based on Statistical Methods for Reliability Data by W.Q. Meeker and L. A. Escobar, Wiley, 1998 and on their class notes. Ramón

More information

Philosophy and Features of the mstate package

Philosophy and Features of the mstate package Introduction Mathematical theory Practice Discussion Philosophy and Features of the mstate package Liesbeth de Wreede, Hein Putter Department of Medical Statistics and Bioinformatics Leiden University

More information

ST745: Survival Analysis: Nonparametric methods

ST745: Survival Analysis: Nonparametric methods ST745: Survival Analysis: Nonparametric methods Eric B. Laber Department of Statistics, North Carolina State University February 5, 2015 The KM estimator is used ubiquitously in medical studies to estimate

More information

A Comparison of Different Approaches to Nonparametric Inference for Subdistributions

A Comparison of Different Approaches to Nonparametric Inference for Subdistributions A Comparison of Different Approaches to Nonparametric Inference for Subdistributions Johannes Mertsching Johannes.Mertsching@gmail.com Master Thesis Supervision: dr. Ronald B. Geskus prof. Chris A.J. Klaassen

More information

Survival Analysis APTS 2016/17. Ingrid Van Keilegom ORSTAT KU Leuven. Glasgow, August 21-25, 2017

Survival Analysis APTS 2016/17. Ingrid Van Keilegom ORSTAT KU Leuven. Glasgow, August 21-25, 2017 Survival Analysis APTS 2016/17 Ingrid Van Keilegom ORSTAT KU Leuven Glasgow, August 21-25, 2017 Basic What is Survival analysis? Survival analysis (or duration analysis) is an area of statistics that and

More information

Multivariate Survival Data With Censoring.

Multivariate Survival Data With Censoring. 1 Multivariate Survival Data With Censoring. Shulamith Gross and Catherine Huber-Carol Baruch College of the City University of New York, Dept of Statistics and CIS, Box 11-220, 1 Baruch way, 10010 NY.

More information

In contrast, parametric techniques (fitting exponential or Weibull, for example) are more focussed, can handle general covariates, but require

In contrast, parametric techniques (fitting exponential or Weibull, for example) are more focussed, can handle general covariates, but require Chapter 5 modelling Semi parametric We have considered parametric and nonparametric techniques for comparing survival distributions between different treatment groups. Nonparametric techniques, such as

More information

Lecture 22 Survival Analysis: An Introduction

Lecture 22 Survival Analysis: An Introduction University of Illinois Department of Economics Spring 2017 Econ 574 Roger Koenker Lecture 22 Survival Analysis: An Introduction There is considerable interest among economists in models of durations, which

More information

Nonparametric two-sample tests of longitudinal data in the presence of a terminal event

Nonparametric two-sample tests of longitudinal data in the presence of a terminal event Nonparametric two-sample tests of longitudinal data in the presence of a terminal event Jinheum Kim 1, Yang-Jin Kim, 2 & Chung Mo Nam 3 1 Department of Applied Statistics, University of Suwon, 2 Department

More information

Modeling Prediction of the Nosocomial Pneumonia with a Multistate model

Modeling Prediction of the Nosocomial Pneumonia with a Multistate model Modeling Prediction of the Nosocomial Pneumonia with a Multistate model M.Nguile Makao 1 PHD student Director: J.F. Timsit 2 Co-Directors: B Liquet 3 & J.F. Coeurjolly 4 1 Team 11 Inserm U823-Joseph Fourier

More information

A multi-state model for the prognosis of non-mild acute pancreatitis

A multi-state model for the prognosis of non-mild acute pancreatitis A multi-state model for the prognosis of non-mild acute pancreatitis Lore Zumeta Olaskoaga 1, Felix Zubia Olaskoaga 2, Marta Bofill Roig 1, Guadalupe Gómez Melis 1 1 Universitat Politècnica de Catalunya

More information

Lecture 2: Martingale theory for univariate survival analysis

Lecture 2: Martingale theory for univariate survival analysis Lecture 2: Martingale theory for univariate survival analysis In this lecture T is assumed to be a continuous failure time. A core question in this lecture is how to develop asymptotic properties when

More information

Lecture 3. Truncation, length-bias and prevalence sampling

Lecture 3. Truncation, length-bias and prevalence sampling Lecture 3. Truncation, length-bias and prevalence sampling 3.1 Prevalent sampling Statistical techniques for truncated data have been integrated into survival analysis in last two decades. Truncation in

More information

Goodness-of-Fit Tests With Right-Censored Data by Edsel A. Pe~na Department of Statistics University of South Carolina Colloquium Talk August 31, 2 Research supported by an NIH Grant 1 1. Practical Problem

More information

Other Survival Models. (1) Non-PH models. We briefly discussed the non-proportional hazards (non-ph) model

Other Survival Models. (1) Non-PH models. We briefly discussed the non-proportional hazards (non-ph) model Other Survival Models (1) Non-PH models We briefly discussed the non-proportional hazards (non-ph) model λ(t Z) = λ 0 (t) exp{β(t) Z}, where β(t) can be estimated by: piecewise constants (recall how);

More information

Censoring and Truncation - Highlighting the Differences

Censoring and Truncation - Highlighting the Differences Censoring and Truncation - Highlighting the Differences Micha Mandel The Hebrew University of Jerusalem, Jerusalem, Israel, 91905 July 9, 2007 Micha Mandel is a Lecturer, Department of Statistics, The

More information

LTCC. Exercises solutions

LTCC. Exercises solutions 1. Markov chain LTCC. Exercises solutions (a) Draw a state space diagram with the loops for the possible steps. If the chain starts in state 4, it must stay there. If the chain starts in state 1, it will

More information

Survival Regression Models

Survival Regression Models Survival Regression Models David M. Rocke May 18, 2017 David M. Rocke Survival Regression Models May 18, 2017 1 / 32 Background on the Proportional Hazards Model The exponential distribution has constant

More information

4. Comparison of Two (K) Samples

4. Comparison of Two (K) Samples 4. Comparison of Two (K) Samples K=2 Problem: compare the survival distributions between two groups. E: comparing treatments on patients with a particular disease. Z: Treatment indicator, i.e. Z = 1 for

More information

Statistical Analysis of Competing Risks With Missing Causes of Failure

Statistical Analysis of Competing Risks With Missing Causes of Failure Proceedings 59th ISI World Statistics Congress, 25-3 August 213, Hong Kong (Session STS9) p.1223 Statistical Analysis of Competing Risks With Missing Causes of Failure Isha Dewan 1,3 and Uttara V. Naik-Nimbalkar

More information

9. Estimating Survival Distribution for a PH Model

9. Estimating Survival Distribution for a PH Model 9. Estimating Survival Distribution for a PH Model Objective: Another Goal of the COX model Estimating the survival distribution for individuals with a certain combination of covariates. PH model assumption:

More information

Survival Analysis. STAT 526 Professor Olga Vitek

Survival Analysis. STAT 526 Professor Olga Vitek Survival Analysis STAT 526 Professor Olga Vitek May 4, 2011 9 Survival Data and Survival Functions Statistical analysis of time-to-event data Lifetime of machines and/or parts (called failure time analysis

More information

Survival analysis in R

Survival analysis in R Survival analysis in R Niels Richard Hansen This note describes a few elementary aspects of practical analysis of survival data in R. For further information we refer to the book Introductory Statistics

More information

Part III. Hypothesis Testing. III.1. Log-rank Test for Right-censored Failure Time Data

Part III. Hypothesis Testing. III.1. Log-rank Test for Right-censored Failure Time Data 1 Part III. Hypothesis Testing III.1. Log-rank Test for Right-censored Failure Time Data Consider a survival study consisting of n independent subjects from p different populations with survival functions

More information

9 Estimating the Underlying Survival Distribution for a

9 Estimating the Underlying Survival Distribution for a 9 Estimating the Underlying Survival Distribution for a Proportional Hazards Model So far the focus has been on the regression parameters in the proportional hazards model. These parameters describe the

More information

Kaplan-Meier in SAS. filename foo url "http://math.unm.edu/~james/small.txt"; data small; infile foo firstobs=2; input time censor; run;

Kaplan-Meier in SAS. filename foo url http://math.unm.edu/~james/small.txt; data small; infile foo firstobs=2; input time censor; run; Kaplan-Meier in SAS filename foo url "http://math.unm.edu/~james/small.txt"; data small; infile foo firstobs=2; input time censor; run; proc print data=small; run; proc lifetest data=small plots=survival;

More information

Multistate models in survival and event history analysis

Multistate models in survival and event history analysis Multistate models in survival and event history analysis Dorota M. Dabrowska UCLA November 8, 2011 Research supported by the grant R01 AI067943 from NIAID. The content is solely the responsibility of the

More information

arxiv: v3 [math.st] 12 Oct 2015

arxiv: v3 [math.st] 12 Oct 2015 Approximative Tests for the Equality of Two Cumulative Incidence Functions of a Competing Risk arxiv:142.229v3 [math.st] 12 Oct 215 Dennis Dobler and Markus Pauly September 28, 218 University of Ulm, Institute

More information

Kernel density estimation in R

Kernel density estimation in R Kernel density estimation in R Kernel density estimation can be done in R using the density() function in R. The default is a Guassian kernel, but others are possible also. It uses it s own algorithm to

More information

Multistate models and recurrent event models

Multistate models and recurrent event models Multistate models Multistate models and recurrent event models Patrick Breheny December 10 Patrick Breheny Survival Data Analysis (BIOS 7210) 1/22 Introduction Multistate models In this final lecture,

More information

A note on the decomposition of number of life years lost according to causes of death

A note on the decomposition of number of life years lost according to causes of death A note on the decomposition of number of life years lost according to causes of death Per Kragh Andersen Research Report 12/2 Department of Biostatistics University of Copenhagen A note on the decomposition

More information

Package Rsurrogate. October 20, 2016

Package Rsurrogate. October 20, 2016 Type Package Package Rsurrogate October 20, 2016 Title Robust Estimation of the Proportion of Treatment Effect Explained by Surrogate Marker Information Version 2.0 Date 2016-10-19 Author Layla Parast

More information

Smoothing the Nelson-Aalen Estimtor Biostat 277 presentation Chi-hong Tseng

Smoothing the Nelson-Aalen Estimtor Biostat 277 presentation Chi-hong Tseng Smoothing the Nelson-Aalen Estimtor Biostat 277 presentation Chi-hong seng Reference: 1. Andersen, Borgan, Gill, and Keiding (1993). Statistical Model Based on Counting Processes, Springer-Verlag, p.229-255

More information

Residuals and model diagnostics

Residuals and model diagnostics Residuals and model diagnostics Patrick Breheny November 10 Patrick Breheny Survival Data Analysis (BIOS 7210) 1/42 Introduction Residuals Many assumptions go into regression models, and the Cox proportional

More information

Continuous Time Markov Chain Examples

Continuous Time Markov Chain Examples Continuous Markov Chain Examples Example Consider a continuous time Markov chain on S {,, } The Markov chain is a model that describes the current status of a match between two particular contestants:

More information

Introduction to cthmm (Continuous-time hidden Markov models) package

Introduction to cthmm (Continuous-time hidden Markov models) package Introduction to cthmm (Continuous-time hidden Markov models) package Abstract A disease process refers to a patient s traversal over time through a disease with multiple discrete states. Multistate models

More information

11 Survival Analysis and Empirical Likelihood

11 Survival Analysis and Empirical Likelihood 11 Survival Analysis and Empirical Likelihood The first paper of empirical likelihood is actually about confidence intervals with the Kaplan-Meier estimator (Thomas and Grunkmeier 1979), i.e. deals with

More information

Survival Analysis I (CHL5209H)

Survival Analysis I (CHL5209H) Survival Analysis Dalla Lana School of Public Health University of Toronto olli.saarela@utoronto.ca January 7, 2015 31-1 Literature Clayton D & Hills M (1993): Statistical Models in Epidemiology. Not really

More information

Analysis of Time-to-Event Data: Chapter 2 - Nonparametric estimation of functions of survival time

Analysis of Time-to-Event Data: Chapter 2 - Nonparametric estimation of functions of survival time Analysis of Time-to-Event Data: Chapter 2 - Nonparametric estimation of functions of survival time Steffen Unkel Department of Medical Statistics University Medical Center Göttingen, Germany Winter term

More information

Multi-state models: a flexible approach for modelling complex event histories in epidemiology

Multi-state models: a flexible approach for modelling complex event histories in epidemiology Multi-state models: a flexible approach for modelling complex event histories in epidemiology Ahmadou Alioum Centre de Recherche INSERM U897 "Epidémiologie et Biostatistique Institut de Santé Publique,

More information

1 Glivenko-Cantelli type theorems

1 Glivenko-Cantelli type theorems STA79 Lecture Spring Semester Glivenko-Cantelli type theorems Given i.i.d. observations X,..., X n with unknown distribution function F (t, consider the empirical (sample CDF ˆF n (t = I [Xi t]. n Then

More information

Chapter 2 - Survival Models

Chapter 2 - Survival Models 2-1 Chapter 2 - Survival Models Section 2.2 - Future Lifetime Random Variable and the Survival Function Let T x = ( Future lifelength beyond age x of an individual who has survived to age x [measured in

More information

Typical Survival Data Arising From a Clinical Trial. Censoring. The Survivor Function. Mathematical Definitions Introduction

Typical Survival Data Arising From a Clinical Trial. Censoring. The Survivor Function. Mathematical Definitions Introduction Outline CHL 5225H Advanced Statistical Methods for Clinical Trials: Survival Analysis Prof. Kevin E. Thorpe Defining Survival Data Mathematical Definitions Non-parametric Estimates of Survival Comparing

More information

Lecture 5 Models and methods for recurrent event data

Lecture 5 Models and methods for recurrent event data Lecture 5 Models and methods for recurrent event data Recurrent and multiple events are commonly encountered in longitudinal studies. In this chapter we consider ordered recurrent and multiple events.

More information

Estimating Load-Sharing Properties in a Dynamic Reliability System. Paul Kvam, Georgia Tech Edsel A. Peña, University of South Carolina

Estimating Load-Sharing Properties in a Dynamic Reliability System. Paul Kvam, Georgia Tech Edsel A. Peña, University of South Carolina Estimating Load-Sharing Properties in a Dynamic Reliability System Paul Kvam, Georgia Tech Edsel A. Peña, University of South Carolina Modeling Dependence Between Components Most reliability methods are

More information

Supply Forecasting via Survival Curve Estimation

Supply Forecasting via Survival Curve Estimation Supply Forecasting via Survival Curve Estimation Gary D. Knott, Ph.D. Civilized Software Inc. 12109 Heritage Park Circle Silver Spring, MD 20906 USA email:csi@civilized.com URL:www.civilized.com Tel: (301)

More information

Journal of Statistical Software

Journal of Statistical Software JSS Journal of Statistical Software January 2011, Volume 38, Issue 2. http://www.jstatsoft.org/ Analyzing Competing Risk Data Using the R timereg Package Thomas H. Scheike University of Copenhagen Mei-Jie

More information

Multistate Modelling Vertical Transmission and Determination of R 0 Using Transition Intensities

Multistate Modelling Vertical Transmission and Determination of R 0 Using Transition Intensities Applied Mathematical Sciences, Vol. 9, 2015, no. 79, 3941-3956 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ams.2015.52130 Multistate Modelling Vertical Transmission and Determination of R 0

More information

I N S T I T U T D E S T A T I S T I Q U E B I O S T A T I S T I Q U E E T S C I E N C E S A C T U A R I E L L E S (I S B A)

I N S T I T U T D E S T A T I S T I Q U E B I O S T A T I S T I Q U E E T S C I E N C E S A C T U A R I E L L E S (I S B A) I N S T I T U T D E S T A T I S T I Q U E B I O S T A T I S T I Q U E E T S C I E N C E S A C T U A R I E L L E S (I S B A) UNIVERSITÉ CATHOLIQUE DE LOUVAIN D I S C U S S I O N P A P E R 153 ESTIMATION

More information

PENALIZED LIKELIHOOD PARAMETER ESTIMATION FOR ADDITIVE HAZARD MODELS WITH INTERVAL CENSORED DATA

PENALIZED LIKELIHOOD PARAMETER ESTIMATION FOR ADDITIVE HAZARD MODELS WITH INTERVAL CENSORED DATA PENALIZED LIKELIHOOD PARAMETER ESTIMATION FOR ADDITIVE HAZARD MODELS WITH INTERVAL CENSORED DATA Kasun Rathnayake ; A/Prof Jun Ma Department of Statistics Faculty of Science and Engineering Macquarie University

More information

Part IV Extensions: Competing Risks Endpoints and Non-Parametric AUC(t) Estimation

Part IV Extensions: Competing Risks Endpoints and Non-Parametric AUC(t) Estimation Part IV Extensions: Competing Risks Endpoints and Non-Parametric AUC(t) Estimation Patrick J. Heagerty PhD Department of Biostatistics University of Washington 166 ISCB 2010 Session Four Outline Examples

More information

Log-linearity for Cox s regression model. Thesis for the Degree Master of Science

Log-linearity for Cox s regression model. Thesis for the Degree Master of Science Log-linearity for Cox s regression model Thesis for the Degree Master of Science Zaki Amini Master s Thesis, Spring 2015 i Abstract Cox s regression model is one of the most applied methods in medical

More information

The Wild Bootstrap for Multivariate Nelson-Aalen Estimators

The Wild Bootstrap for Multivariate Nelson-Aalen Estimators arxiv:1602.02071v2 [stat.me] 3 Feb 2017 The Wild Bootstrap for Multivariate Nelson-Aalen Estimators Tobias Bluhmki, Dennis Dobler, Jan Beyersmann, Markus Pauly Ulm University, Institute of Statistics,

More information

Duration Analysis. Joan Llull

Duration Analysis. Joan Llull Duration Analysis Joan Llull Panel Data and Duration Models Barcelona GSE joan.llull [at] movebarcelona [dot] eu Introduction Duration Analysis 2 Duration analysis Duration data: how long has an individual

More information

MAS3301 / MAS8311 Biostatistics Part II: Survival

MAS3301 / MAS8311 Biostatistics Part II: Survival MAS3301 / MAS8311 Biostatistics Part II: Survival M. Farrow School of Mathematics and Statistics Newcastle University Semester 2, 2009-10 1 13 The Cox proportional hazards model 13.1 Introduction In the

More information

SMSTC (2007/08) Probability.

SMSTC (2007/08) Probability. SMSTC (27/8) Probability www.smstc.ac.uk Contents 12 Markov chains in continuous time 12 1 12.1 Markov property and the Kolmogorov equations.................... 12 2 12.1.1 Finite state space.................................

More information

ST745: Survival Analysis: Parametric

ST745: Survival Analysis: Parametric ST745: Survival Analysis: Parametric Eric B. Laber Department of Statistics, North Carolina State University January 13, 2015 ...the statistician knows... that in nature there never was a normal distribution,

More information

The coxvc_1-1-1 package

The coxvc_1-1-1 package Appendix A The coxvc_1-1-1 package A.1 Introduction The coxvc_1-1-1 package is a set of functions for survival analysis that run under R2.1.1 [81]. This package contains a set of routines to fit Cox models

More information

A multistate additive relative survival semi-markov model

A multistate additive relative survival semi-markov model 46 e Journées de Statistique de la SFdS A multistate additive semi-markov Florence Gillaizeau,, & Etienne Dantan & Magali Giral, & Yohann Foucher, E-mail: florence.gillaizeau@univ-nantes.fr EA 4275 - SPHERE

More information

Survival Times (in months) Survival Times (in months) Relative Frequency. Relative Frequency

Survival Times (in months) Survival Times (in months) Relative Frequency. Relative Frequency Smooth Goodness-of-Fit Tests in Hazard-Based Models by Edsel A. Pe~na Department of Statistics University of South Carolina at Columbia E-Mail: pena@stat.sc.edu Univ. of Virginia Colloquium Talk Department

More information

Survival Analysis for Case-Cohort Studies

Survival Analysis for Case-Cohort Studies Survival Analysis for ase-ohort Studies Petr Klášterecký Dept. of Probability and Mathematical Statistics, Faculty of Mathematics and Physics, harles University, Prague, zech Republic e-mail: petr.klasterecky@matfyz.cz

More information

Pricing and Risk Analysis of a Long-Term Care Insurance Contract in a non-markov Multi-State Model

Pricing and Risk Analysis of a Long-Term Care Insurance Contract in a non-markov Multi-State Model Pricing and Risk Analysis of a Long-Term Care Insurance Contract in a non-markov Multi-State Model Quentin Guibert Univ Lyon, Université Claude Bernard Lyon 1, ISFA, Laboratoire SAF EA2429, F-69366, Lyon,

More information

CDA5530: Performance Models of Computers and Networks. Chapter 3: Review of Practical

CDA5530: Performance Models of Computers and Networks. Chapter 3: Review of Practical CDA5530: Performance Models of Computers and Networks Chapter 3: Review of Practical Stochastic Processes Definition Stochastic ti process X = {X(t), t T} is a collection of random variables (rvs); one

More information

Survival Times (in months) Survival Times (in months) Relative Frequency. Relative Frequency

Survival Times (in months) Survival Times (in months) Relative Frequency. Relative Frequency Smooth Goodness-of-Fit Tests in Hazard-Based Models by Edsel A. Pe~na Department of Statistics University of South Carolina at Columbia E-Mail: pena@stat.sc.edu Univ. of Georgia Colloquium Talk Department

More information