Lecture 3. Truncation, length-bias and prevalence sampling

Size: px
Start display at page:

Download "Lecture 3. Truncation, length-bias and prevalence sampling"

Transcription

1 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 survival analysis refers to an incomplete data mechanism. It describes a sampling constraint that a failure time variable is observable only if it falls in a certain region. When the value of failure time falls outside the region, the information about the variable is completely lost and therefore excluded from the data set. Truncated survival data typically arise when prevalent sampling is used for recruiting cohort subjects.

2 When conducting a natural history study, the incident and prevalent samplings are commonly used for recruiting cohort subjects. Incident cohort. An incident cohort is identified by randomly sampling subjects with initial events (time origin 0) occurring within a calendar time interval. The subjects are then followed until occurrence of the failure event, loss to follow-up or end-of-study. The data collected from an incident cohort are the typical survival data subject to right censoring.

3 Prevalent cohort. When failure times are long, the incident cohort design is inefficient for natural history studies because it usually takes a long follow-up time to observe enough failure events. In contrast, a prevalent sampling design which draws samples from a disease prevalent population is more focused and thus more practical in real studies. The prevalent sample is formed by subjects whose initial events had occurred but have not experienced the failure event at the time of recruitment, τ.

4 :27:01 1/1 Unnamed Page (#27)

5 The prevalent sampling can be described using a distributional truncation model I, or a non-stationary Poisson process model II: I. Define T 0 as the time from the disease incidence to the failure event for subjects who became diseased in a calendar time interval [0, τ], τ > 0. The variable W 0 is the time from disease incidence to the (potential) recruitment time, termed as the left truncation time. Under the prevalent sampling, the probability density of the observed (W, T ) is the population probability density of (W 0, T 0 ) given T 0 W 0 : p W,T (w, t) = p W 0,T 0 T 0 W 0 (w, t T 0 W 0 ).

6 II. Let T 0 be the time from the disease incidence to the failure event. Assume the initial events occur over the calendar time as a non-stationary Poisson process with intensity/rate λ(u 0 ), u 0 [0, τ], τ > 0. The prevalent sampling population includes those {(u 0, t 0 )} where the disease incidence occurs within calendar time interval [0, τ] and the failure event occurs after calendar time τ, i.e., u 0 + t 0 τ.

7 Remark: In this lecture (W 0, T 0 ) represent population variables and (W, T ) represent observed variables subject to prevalent sampling.

8 Models I and II are connected through the distributions of disease incidence and failure times. In Model I, - g is the marginal density function of W 0 - S u (t) = Pr(T 0 t W 0 = u) is the survival function of T 0 given W 0 = u - when T 0 is independent of W 0, S u (t) = S(t) = Pr(T 0 t) In Model II, define the pdf g 0 (u) = λ(u)/ τ λ(v)dv as the normalized 0 rate for u [0, τ], the truncation time as W 0 = τ U 0, and g(w 0 ) = g 0 (τ w 0 ).

9 Summary: - Under Model II, conditioning on the number of initial events occurring in [0, τ], the unordered truncation times {W 0 } can be considered as a set of iid random variables with pdf g, and the corresponding set of bivariate vectors {(W 0, T 0 )} form a set of iid bivariate random vectors. The prevalent population includes individuals with {(w 0, t 0 )} where the failure event occurs after calendar time τ, i.e., t 0 w 0. Use the prevalent sampling to collect untruncated data until n pairs of untruncated (W 1, T 1 ),..., (W n, T n ) are observed. - Under either Model I or Model II, the prevalent sample is then formed by n pairs of iid untruncated (W 1, T 1 ),..., (W n, T n ).

10 The independent truncation assumption refers to the assumption that T 0 and W 0 are independent of each other. In application this independent truncation condition is satisfied when the failure time is independent of the calendar time of incidence of the initiating event. The joint density of the observed (W, T ) can then be expressed as p W,T (w, t) = g(w)f(t)i(t w) P(T 0 W 0 ) = g(w)f(t)i(t w) S(u)g(u)du (1)

11 Example. Suppose a random sample of women with breast cancer (b.c.) are recruited for observation of survival. The failure time T 0 is defined as the time from onset of b.c. to death. Suppose the time of recruitment, τ, is a fixed calendar time. Then, g(w) can be interpreted as the occurrence density of b.c. at calendar time τ w. The independent truncation condition is satisfied when the failure time T 0 (residual lifetime for b.c.) is independent of the calendar time of incidence of b.c.

12 3.2 Length-biased sampling Length-biased sampling could arise in many epidemiological studies when survival data are collected from a disease population. Assume (i) The rate of occurrence of incidence of disease, λ(u), remains constant over time. (ii) The probability distribution of the failure time T 0 is independent of when the incidence of disease occurred. Conditions in (i) and (ii) together refer to as the equilibrium state. The two conditions define the so-called stable diseases.

13 The breast cancer example. In the breast cancer (b.c.) example, the prevalent sampling is length-biased if (i) the rate of occurrence of b.c., λ( ), remains constant over time, and (ii) the density function of the time from b.c. to death, f, is independent of when b.c. occurred.

14 For simplicity of discussion, assume the failure time T 0 has finite support [0, τ]. When conditions (i) and (ii) are satisfied, the truncation time T 0 follows uniform distribution over [0, τ] - note that this condition would be satisfied if the disease incidence rate has been constant for a long time ( τ units) before the prevalent sample is drawn. The joint density of observed (W, T ) in (1) becomes p W,T (w, t) = f(t)i(t w) 0 S(u)du = f(t)i(t w) µ, where and µ = E[T 0 ].

15 Define the forward failure time as R = T W. The joint density function of (W, V ) can be derived as p W,R (w, r) = f(w + r)i(w 0, r 0) µ, (2) which is known in renewal theory the joint density function of backward and forward recurrence times (Feller, 1971). In general, treating length-biased data as the usual data would lead to biased analytical results. The amount of bias could be large or small, depending on models and problems.

16 Length bias: a few marginal distributions The sampling density functions of T, W and R can be derived, based on (2), as p T (t) = tf(t)i(t 0)/µ, p W (w) = S(w)I(w 0)/µ, p R (r) = S(r)I(r 0)/µ. The distribution of (3) is generally referred to as the length-biased distribution. Exercise: Drive the marginal pdf s of T, W and R from the joint pdf p W,R (w, r).

17 The survivorship function S can be estimated by a weighted empirical distribution function, with weight inversely proportional to t i : Ŝ n (t) = n 1 ˆµ n {t 1 i I(t i t)}, i=1 where ˆµ = {n 1 j t 1 j } 1 serves as an appropriate estimate of µ, since n 1 j t 1 j estimates µ 1. The estimator Ŝn was first proposed by Cox (1969, Book) and can be proven to be the nonparametric maximum likelihood estimator of S, a special case under Vardi s selection bias models (1982, 1985, AS).

18 Exercise: Identify assumptions under which ˆµ converges to µ in probability. Also prove the convergence result.

19 Density estimation for length-biased data. Following the same weighting procedure, a kernel estimator of the density function f can be derived as ˆf n (t) = n 1 ˆµ n {t 1 i K h (t t i )}, i=1 where K h (x) = h 1 K(h 1 x), h > 0, with K a kernel function. Alternatively, one could first estimate the length-biased density, (2), by an ordinary kernel estimator and then use the relationship of (2) and f to obtain an estimator of f.

20 PHM for length-biased data. Under the proportional hazards model, for uncensored length-biased data, a risk set sampling technique can be adopted for estimating regression parameters. For t j t i, let j (t i ) be a binary variable which equals 1 with probability t i /t j, and 0 with probability 1 (t i /t j ). The indicators j (t i ) are used to identify bias-adjusted risk sets and to construct pseudo-likelihood equations. Regression parameter estimates are then derived by solving the score equations. For censored length-biased data, statistical methods and theory have been further developed in last few years for non- and semi-parametric models.

21 Prevalence = Incidence Duration Use Model II. Recall that λ(u) is the disease incidence rate at the calendar time u, and S u is the survival function of T 0 for those patients whose disease was initiated at u. The disease prevalence rate at calendar time τ can be obtained as P (τ) = τ λ(u)s u(τ u)du. When the equilibrium condition is satisfied, the incidence rate is a constant (λ(u) = λ 0 ) and the survival function is independent of u (S u = S), and P (τ) = λ 0 τ S(τ u)du = λ 0 S(u)du = λ 0 µ is functionally independent of τ. Thus, let P (τ) = p 0 and we derive p 0 = λ 0 µ (Prevalence = Incidence duration). 0

22 Length bias: full and conditional likelihoods Data: iid (w 1, t 1 ),..., (w n, t n ) L m : density function of {t i } L c : density function of {w i } conditioning on {t i } The full likelihood is L = L m L c = { n i=1 } { t i f(t i ) n µ i=1 } 1 By the factorization theorem, the observed failure times {t i } serve as sufficient statistics for f in cases that f is either a parametric or nonparametric function. Thus, conditioning on {t i }, the truncation times {w i } do not contain additional information for f. t i

23 3.3 Left truncation Length-biased data can be viewed as a special case of left truncated data, since the conditional density of the observed t i given w i is f(t i )I(t i w i ) S(w i ) (3) which corresponds to the density function of left truncated failure time. By viewing length-biased data as left truncated data, we next consider how to analyze left truncated data in a general setting. It is important to indicate that the validity of the truncated density in (3) depends only on the independent truncation assumption, (ii), and not on the constant incidence rate assumption, (i).

24 Notation i = 1, 2,..., n: index for subjects (W 1, T 1 ),..., (W n, T n ): iid observations R(t) = {(W j, T j ) : W j t T j, j = 1, 2,..., n}: risk set at t Y i (t) = I( W j t T j ): at-risk indicator Y (t) = n i=1 Y i(t): total no. of subjects at risk at t N i (t) = I(T i t): count subject i s failure event prior to time t N(t) = n i=1 N i(t): total number of observed events prior to time t

25 The conditional likelihood based on {(w i, t i )} given {w i } can be written as L c = n i=1 {f(t i)/s(w i )}. Let t (1) <... < t (J) be the distinct and ordered values of t 1,..., t n and let w (1),..., w (J) be the corresponding truncation times. By assigning probability mass to and only to {t (j) }, we can write f(t (j) ) S(w (j) ) = S(t (k j+1)) S(t (kj)) S(t (k j+2)) S(t (kj+1))... f(t (j)) S(t (j) ) = (1 λ(t (kj))) (1 λ(t (kj+1)))... λ(t (j) ) (4) where t (kj) satisfies t (kj 1) < w (j) t (kj). Note that (w (j), t (j) ) falls in the risk sets at t (kj), t (kj+1),... and t (j), and that the factorization in (4) consists of both the factor λ(t (j) ) at t (j), and the factor 1 λ(t (k) ) if (w (j), t (j) ) falls in the risk set at t (k).

26 The risk set at t is R(t) = {(w i, t i ) : w i t t i }, and d (j) the number of failures at t (j), Y (t (j) ) as the number of subjects in R(t (j) ). The conditional likelihood L c can be reparameterized as L c = = J [ ] dn(t(j) ) [ f(t(j) ) S(t(j+1) ) S(t (j) ) S(t (j) ) J [ ] dn(t(j) ) [ ] Y (t(j) ) dn(t λ(j) 1 (j) ) λ(j) j=1 j=1 ] Y (t(j) ) dn(t (j) )

27 Thus, following an argument similar to the one in Kaplan and Meier (1958), the unique mle of λ (j) can be obtained as dn(t (j) )/Y (t (j) ) and the product-limit estimator Ŝ(t) = t (j) < t ( 1 dn(t ) (j)) Y (t (j) ) (5) is the unique nonparametric MLE of L c. This is also called the truncation product-limit estimator. Exercise: Does the existence of NPMLE require conditions?

28 Example Data: (4, 5), (0, 4), (5, 7), (1, 2), (2, 8), (1, 5) t (i) dn(t (i) ) Y (t (i) ) R(t (1) ) = {(0, 4), (1, 2), (2, 8), (1, 5)} R(t (2) ) = {(4, 5), (0, 4), (2, 8), (1, 5)} The truncation product-limit estimate is thus Ŝ(2) = ( 1 Ŝ(4) = 1 1 ) = 3 4 4

29 ( Ŝ(5) = 1 1 ) ( 1 1 ) = ( 4 Ŝ(7) = 1 1 ) ( 1 1 ) ( 1 2 ) = ( 4 Ŝ(8) = 1 1 ) ( 1 1 ) ( 1 2 ) ( 1 1 ) = Ŝ(8 + ) = 0 Note: Unlike right censored data, risk sets for truncated data usually are not nested.

30 3.4 Left Truncated and Right Censored Data For i = 1,..., n, assume the following independent censoring condition: Conditional on the observed W i = w i, the time from τ i to the failure event, R i, is independent of the time from τ i to censoring, D i. This independent censoring condition does not, however, imply independence between the biased failure time T i (= W i + R i ) and the censoring time C i (= W i + D i ). Let x i = min{t i, c i } be the time from the initiating event to censoring, and δ i = I(x i = t i ) the censoring indicator. Conditional on w i, under the independent censoring condition, the density of (w i, x i, δ i ) is proportional to f(x i ) δi S(x i ) 1 δi /S(w i ).

31 :33:37 1/1 Unnamed Page (#25)

32 Notation i = 1, 2,..., n: index for subjects (W 1, X 1, 1 ),..., (W n, X n, n ): iid observations R(t) = {(W j, X j, j ) : W j t X j, j = 1, 2,..., n}: risk set at t Y i (t) = I( W j t X j ): at-risk indicator Y (t) = n i=1 Y i(t): total no. of subjects at risk at t N i (t) = I(X i t, j = 1): count subject i s failure event prior to time t N(t) = n i=1 N i(t): total number of observed events prior to time t

33 Given a sample of independent and identically distributed observations (w 1, x 1, δ 1 ),..., (w n, x n, δ n ), the full likelihood is L = L c L m, where L c is the conditional likelihood corresponding to the density of {(w, x, δ)} given {w}, and L m is the marginal likelihood of {w}. The full likelihood function can be expressed as L = L{ c (S, S D ) L m (S, G) n } f(x i ) δi S(x i ) 1 δi f D W (x i w i w i ) 1 δi S D W (x i w i w i ) δi = S(w i=1 i ) { n } S(w i )g(w i ) i=1 S(w)g(w)dw { n } { n } f(x i ) δi S(x i ) 1 δi S(w i )g(w i ). S(w i ) S(w)g(w)dw i=1 i=1

34 Thus L c n { } f(xi ) δi S(x i ) 1 δi i=1 S(w i ) L m = n i=1 { } S(wi )g(w i ) S(w)g(w)dw Statistical approaches based on the conditional likelihood, L c, are considered methods for left truncated and right censored data.

35 Estimation of S: Using techniques similar to the previous one, the product-limit estimator Ŝ(t) = x (j) < t ( 1 dn(x ) (j)) Y (x (j) ) (6) can be derived as the unique nonparametric MLE from the conditional likelihood L c. This is also called the truncation product-limit estimator.

36 Estimation of G: By assigning probability to, and only to {w i }, the maximum value of the marginal likelihood L m (S, G) is the same as the maximum value of a multinomial likelihood, where the unique maximum likelihood estimator is the empirical distribution. With S fixed as Ŝ (the conditional MLE), the maximum value of L m (Ŝ, G) is achieved by the inverse weighting estimator: Ĝ(w; Ŝ) = n i=1 Ŝ 1 (w i )I(w i w) n i=1 Ŝ 1 (w i ). The NPMLEs from the full likelihood L can then be derived as (Ŝ, Ĝ( ; Ŝ)).

37 Risk-set-based methods. For left truncated and right censored data, we can used the revised risk sets to derive modified Greenwoods Formula - it still holds for the estimation of the asymptotic variance of the truncation product-limit estimator. modified partial likelihood method - it still holds for the estimation of β in the proportional hazards model. modified log-rank tests - it still holds for testing the difference between two groups.

38 Remarks: 1. Essentially, censoring and truncation share some significant similarities in statistical analysis - especially, the similarities in risk-set-based methods. 2. Nevertheless, there exists significant differences that are not all emphasized in this course. In general, censoring is a nuisance distribution but truncation distribution could be important itself (e.g. HIV infection distribution over calendar time) or consist of information about the failure time distribution.

39 Example. For length-biased failure time data (without censoring), the weighted estimator of S, Ŝn(t) = n 1 ˆµ n i=1 {t 1 i I(t i t)}, is much more efficient that the product-limit estimator. Example. Knowing truncation distribution completely or partially can substantially improve estimation for distribution of T 0. Note that for censored survival data the knowledge of censoring distribution does not improve estimation of the survival distribution. Some on-going research...

40 Exercise. Suppose a bivariate random vector (T, Y ) has the joint pdf p T,Y (t, y) = f(y)i(y t 0)/µ, where f is a continuous pdf for a positive-valued random variable and µ = yf(y)dy (mean based on f). Assume the moments based on f are finite up to the 4th order. Suppose (T 1, Y 1 ),..., (T n, Y n ) are iid random vectors with the same pdf of (T, Y ). Part I. (a) Show that the marginal pdf of Y is yf(y)/µ (length-bias distribution) and show that the pdf of Y conditioning on T = t is f(y)i(y t)/(1 F (t)). (b) Find the expected value of 1/Y. Is E[Y ] less than, equal to, or greater than µ? (c) Does n 1 n i=1 {1/Y i} converge to µ 1 in probability? Explain. (d) Derive an asymptotically normally distributed estimator of µ and describe the estimator s asymptotic distribution.

41 Part II. Now assume f is the Gamma(γ, θ) density function with known γ > 0 and unknown θ > 0, where f(y) = θ γ y γ 1 e θy /Γ(γ). (e) Show that (Y 1,..., Y n ) is a sufficient statistic for this family of distribution {p T,Y }. Does your conclusion depend on the choice of Gamma distribution? (That is, does your conclusion hold if the Gamma distribution model is replaced by a different distribution model?) (f) Based on (t 1, y 1 ),..., (t n, y n ), find the mle of θ. Find the asymptotic distribution for the mle of θ. (g) What is the mle of µ?

42 Part III. In this part we assume f is the pdf of an exponential distribution: f(y; θ) = θe θy I(y > 0), θ > 0 (h) Consider the full likelihood L, based on the pdf of observed {(t 1, y 1 ),..., (t n, y n )}. Calculate the Fisher information for θ based on L. (i) Consider the conditional likelihood L c, based on the pdf of observed {(t 1, y 1 ),..., (t n, y n )} given observed {t 1,,..., t n }. Derive the MLE of θ from L c and calculate the Fisher information for θ based on L c. (j) Compare the Fisher informations for θ derived from L and L c. Please provide a discussion of your result.

43 Remarks: 1. For Gamma(γ, θ) distribution, the mean is µ = γ/θ and the variance is γ/θ If n is a positive integer, then the Gamma function has the property that Γ(n) = (n 1)!.

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

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

Survival Distributions, Hazard Functions, Cumulative Hazards

Survival Distributions, Hazard Functions, Cumulative Hazards BIO 244: Unit 1 Survival Distributions, Hazard Functions, Cumulative Hazards 1.1 Definitions: The goals of this unit are to introduce notation, discuss ways of probabilistically describing the distribution

More information

STAT331. Cox s Proportional Hazards Model

STAT331. Cox s Proportional Hazards Model STAT331 Cox s Proportional Hazards Model In this unit we introduce Cox s proportional hazards (Cox s PH) model, give a heuristic development of the partial likelihood function, and discuss adaptations

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

Part III Measures of Classification Accuracy for the Prediction of Survival Times

Part III Measures of Classification Accuracy for the Prediction of Survival Times Part III Measures of Classification Accuracy for the Prediction of Survival Times Patrick J Heagerty PhD Department of Biostatistics University of Washington 102 ISCB 2010 Session Three Outline Examples

More information

UNIVERSITY OF CALIFORNIA, SAN DIEGO

UNIVERSITY OF CALIFORNIA, SAN DIEGO UNIVERSITY OF CALIFORNIA, SAN DIEGO Estimation of the primary hazard ratio in the presence of a secondary covariate with non-proportional hazards An undergraduate honors thesis submitted to the Department

More information

Estimation of Conditional Kendall s Tau for Bivariate Interval Censored Data

Estimation of Conditional Kendall s Tau for Bivariate Interval Censored Data Communications for Statistical Applications and Methods 2015, Vol. 22, No. 6, 599 604 DOI: http://dx.doi.org/10.5351/csam.2015.22.6.599 Print ISSN 2287-7843 / Online ISSN 2383-4757 Estimation of Conditional

More information

Survival Analysis. Lu Tian and Richard Olshen Stanford University

Survival Analysis. Lu Tian and Richard Olshen Stanford University 1 Survival Analysis Lu Tian and Richard Olshen Stanford University 2 Survival Time/ Failure Time/Event Time We will introduce various statistical methods for analyzing survival outcomes What is the survival

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

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

ST495: Survival Analysis: Maximum likelihood

ST495: Survival Analysis: Maximum likelihood ST495: Survival Analysis: Maximum likelihood Eric B. Laber Department of Statistics, North Carolina State University February 11, 2014 Everything is deception: seeking the minimum of illusion, keeping

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

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

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

Multivariate Survival Analysis

Multivariate Survival Analysis Multivariate Survival Analysis Previously we have assumed that either (X i, δ i ) or (X i, δ i, Z i ), i = 1,..., n, are i.i.d.. This may not always be the case. Multivariate survival data can arise in

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

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

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

Nonparametric rank based estimation of bivariate densities given censored data conditional on marginal probabilities

Nonparametric rank based estimation of bivariate densities given censored data conditional on marginal probabilities Hutson Journal of Statistical Distributions and Applications (26 3:9 DOI.86/s4488-6-47-y RESEARCH Open Access Nonparametric rank based estimation of bivariate densities given censored data conditional

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

Power and Sample Size Calculations with the Additive Hazards Model

Power and Sample Size Calculations with the Additive Hazards Model Journal of Data Science 10(2012), 143-155 Power and Sample Size Calculations with the Additive Hazards Model Ling Chen, Chengjie Xiong, J. Philip Miller and Feng Gao Washington University School of Medicine

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

Likelihood Construction, Inference for Parametric Survival Distributions

Likelihood Construction, Inference for Parametric Survival Distributions Week 1 Likelihood Construction, Inference for Parametric Survival Distributions In this section we obtain the likelihood function for noninformatively rightcensored survival data and indicate how to make

More information

Modelling geoadditive survival data

Modelling geoadditive survival data Modelling geoadditive survival data Thomas Kneib & Ludwig Fahrmeir Department of Statistics, Ludwig-Maximilians-University Munich 1. Leukemia survival data 2. Structured hazard regression 3. Mixed model

More information

Fall 2017 STAT 532 Homework Peter Hoff. 1. Let P be a probability measure on a collection of sets A.

Fall 2017 STAT 532 Homework Peter Hoff. 1. Let P be a probability measure on a collection of sets A. 1. Let P be a probability measure on a collection of sets A. (a) For each n N, let H n be a set in A such that H n H n+1. Show that P (H n ) monotonically converges to P ( k=1 H k) as n. (b) For each n

More information

Semiparametric Regression

Semiparametric Regression Semiparametric Regression Patrick Breheny October 22 Patrick Breheny Survival Data Analysis (BIOS 7210) 1/23 Introduction Over the past few weeks, we ve introduced a variety of regression models under

More information

Chapter 2 Inference on Mean Residual Life-Overview

Chapter 2 Inference on Mean Residual Life-Overview Chapter 2 Inference on Mean Residual Life-Overview Statistical inference based on the remaining lifetimes would be intuitively more appealing than the popular hazard function defined as the risk of immediate

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

Support Vector Hazard Regression (SVHR) for Predicting Survival Outcomes. Donglin Zeng, Department of Biostatistics, University of North Carolina

Support Vector Hazard Regression (SVHR) for Predicting Survival Outcomes. Donglin Zeng, Department of Biostatistics, University of North Carolina Support Vector Hazard Regression (SVHR) for Predicting Survival Outcomes Introduction Method Theoretical Results Simulation Studies Application Conclusions Introduction Introduction For survival data,

More information

GOODNESS-OF-FIT TESTS FOR ARCHIMEDEAN COPULA MODELS

GOODNESS-OF-FIT TESTS FOR ARCHIMEDEAN COPULA MODELS Statistica Sinica 20 (2010), 441-453 GOODNESS-OF-FIT TESTS FOR ARCHIMEDEAN COPULA MODELS Antai Wang Georgetown University Medical Center Abstract: In this paper, we propose two tests for parametric models

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

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

Product-limit estimators of the gap time distribution of a renewal process under different sampling patterns

Product-limit estimators of the gap time distribution of a renewal process under different sampling patterns Product-limit estimators of the gap time distribution of a renewal process under different sampling patterns arxiv:13.182v1 [stat.ap] 28 Feb 21 Richard D. Gill Department of Mathematics University of Leiden

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

Key Words: survival analysis; bathtub hazard; accelerated failure time (AFT) regression; power-law distribution.

Key Words: survival analysis; bathtub hazard; accelerated failure time (AFT) regression; power-law distribution. POWER-LAW ADJUSTED SURVIVAL MODELS William J. Reed Department of Mathematics & Statistics University of Victoria PO Box 3060 STN CSC Victoria, B.C. Canada V8W 3R4 reed@math.uvic.ca Key Words: survival

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

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

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

1 Delayed Renewal Processes: Exploiting Laplace Transforms

1 Delayed Renewal Processes: Exploiting Laplace Transforms IEOR 6711: Stochastic Models I Professor Whitt, Tuesday, October 22, 213 Renewal Theory: Proof of Blackwell s theorem 1 Delayed Renewal Processes: Exploiting Laplace Transforms The proof of Blackwell s

More information

Statistical Methods for Alzheimer s Disease Studies

Statistical Methods for Alzheimer s Disease Studies Statistical Methods for Alzheimer s Disease Studies Rebecca A. Betensky, Ph.D. Department of Biostatistics, Harvard T.H. Chan School of Public Health July 19, 2016 1/37 OUTLINE 1 Statistical collaborations

More information

Probability Distributions Columns (a) through (d)

Probability Distributions Columns (a) through (d) Discrete Probability Distributions Columns (a) through (d) Probability Mass Distribution Description Notes Notation or Density Function --------------------(PMF or PDF)-------------------- (a) (b) (c)

More information

FULL LIKELIHOOD INFERENCES IN THE COX MODEL

FULL LIKELIHOOD INFERENCES IN THE COX MODEL October 20, 2007 FULL LIKELIHOOD INFERENCES IN THE COX MODEL BY JIAN-JIAN REN 1 AND MAI ZHOU 2 University of Central Florida and University of Kentucky Abstract We use the empirical likelihood approach

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

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

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

Censoring mechanisms

Censoring mechanisms Censoring mechanisms Patrick Breheny September 3 Patrick Breheny Survival Data Analysis (BIOS 7210) 1/23 Fixed vs. random censoring In the previous lecture, we derived the contribution to the likelihood

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

STAT 6385 Survey of Nonparametric Statistics. Order Statistics, EDF and Censoring

STAT 6385 Survey of Nonparametric Statistics. Order Statistics, EDF and Censoring STAT 6385 Survey of Nonparametric Statistics Order Statistics, EDF and Censoring Quantile Function A quantile (or a percentile) of a distribution is that value of X such that a specific percentage of the

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

Simple and Fast Overidentified Rank Estimation for Right-Censored Length-Biased Data and Backward Recurrence Time

Simple and Fast Overidentified Rank Estimation for Right-Censored Length-Biased Data and Backward Recurrence Time Biometrics 74, 77 85 March 2018 DOI: 10.1111/biom.12727 Simple and Fast Overidentified Rank Estimation for Right-Censored Length-Biased Data and Backward Recurrence Time Yifei Sun, 1, * Kwun Chuen Gary

More information

Continuous case Discrete case General case. Hazard functions. Patrick Breheny. August 27. Patrick Breheny Survival Data Analysis (BIOS 7210) 1/21

Continuous case Discrete case General case. Hazard functions. Patrick Breheny. August 27. Patrick Breheny Survival Data Analysis (BIOS 7210) 1/21 Hazard functions Patrick Breheny August 27 Patrick Breheny Survival Data Analysis (BIOS 7210) 1/21 Introduction Continuous case Let T be a nonnegative random variable representing the time to an event

More information

Analysing geoadditive regression data: a mixed model approach

Analysing geoadditive regression data: a mixed model approach Analysing geoadditive regression data: a mixed model approach Institut für Statistik, Ludwig-Maximilians-Universität München Joint work with Ludwig Fahrmeir & Stefan Lang 25.11.2005 Spatio-temporal regression

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

Multistate models and recurrent event models

Multistate models and recurrent event models and recurrent event models Patrick Breheny December 6 Patrick Breheny University of Iowa Survival Data Analysis (BIOS:7210) 1 / 22 Introduction In this final lecture, we will briefly look at two other

More information

Simple techniques for comparing survival functions with interval-censored data

Simple techniques for comparing survival functions with interval-censored data Simple techniques for comparing survival functions with interval-censored data Jinheum Kim, joint with Chung Mo Nam jinhkim@suwon.ac.kr Department of Applied Statistics University of Suwon Comparing survival

More information

Step-Stress Models and Associated Inference

Step-Stress Models and Associated Inference Department of Mathematics & Statistics Indian Institute of Technology Kanpur August 19, 2014 Outline Accelerated Life Test 1 Accelerated Life Test 2 3 4 5 6 7 Outline Accelerated Life Test 1 Accelerated

More information

Quantile Regression for Residual Life and Empirical Likelihood

Quantile Regression for Residual Life and Empirical Likelihood Quantile Regression for Residual Life and Empirical Likelihood Mai Zhou email: mai@ms.uky.edu Department of Statistics, University of Kentucky, Lexington, KY 40506-0027, USA Jong-Hyeon Jeong email: jeong@nsabp.pitt.edu

More information

Lecture 1. Introduction Statistics Statistical Methods II. Presented January 8, 2018

Lecture 1. Introduction Statistics Statistical Methods II. Presented January 8, 2018 Introduction Statistics 211 - Statistical Methods II Presented January 8, 2018 linear models Dan Gillen Department of Statistics University of California, Irvine 1.1 Logistics and Contact Information Lectures:

More information

On consistency of Kendall s tau under censoring

On consistency of Kendall s tau under censoring Biometria (28), 95, 4,pp. 997 11 C 28 Biometria Trust Printed in Great Britain doi: 1.193/biomet/asn37 Advance Access publication 17 September 28 On consistency of Kendall s tau under censoring BY DAVID

More information

Efficient Semiparametric Estimators via Modified Profile Likelihood in Frailty & Accelerated-Failure Models

Efficient Semiparametric Estimators via Modified Profile Likelihood in Frailty & Accelerated-Failure Models NIH Talk, September 03 Efficient Semiparametric Estimators via Modified Profile Likelihood in Frailty & Accelerated-Failure Models Eric Slud, Math Dept, Univ of Maryland Ongoing joint project with Ilia

More information

Computational treatment of the error distribution in nonparametric regression with right-censored and selection-biased data

Computational treatment of the error distribution in nonparametric regression with right-censored and selection-biased data Computational treatment of the error distribution in nonparametric regression with right-censored and selection-biased data Géraldine Laurent 1 and Cédric Heuchenne 2 1 QuantOM, HEC-Management School of

More information

COPYRIGHTED MATERIAL CONTENTS. Preface Preface to the First Edition

COPYRIGHTED MATERIAL CONTENTS. Preface Preface to the First Edition Preface Preface to the First Edition xi xiii 1 Basic Probability Theory 1 1.1 Introduction 1 1.2 Sample Spaces and Events 3 1.3 The Axioms of Probability 7 1.4 Finite Sample Spaces and Combinatorics 15

More information

ST5212: Survival Analysis

ST5212: Survival Analysis ST51: Survival Analysis 8/9: Semester II Tutorial 1. A model for lifetimes, with a bathtub-shaped hazard rate, is the exponential power distribution with survival fumction S(x) =exp{1 exp[(λx) α ]}. (a)

More information

f X (x) = λe λx, , x 0, k 0, λ > 0 Γ (k) f X (u)f X (z u)du

f X (x) = λe λx, , x 0, k 0, λ > 0 Γ (k) f X (u)f X (z u)du 11 COLLECTED PROBLEMS Do the following problems for coursework 1. Problems 11.4 and 11.5 constitute one exercise leading you through the basic ruin arguments. 2. Problems 11.1 through to 11.13 but excluding

More information

Stat 5101 Lecture Notes

Stat 5101 Lecture Notes Stat 5101 Lecture Notes Charles J. Geyer Copyright 1998, 1999, 2000, 2001 by Charles J. Geyer May 7, 2001 ii Stat 5101 (Geyer) Course Notes Contents 1 Random Variables and Change of Variables 1 1.1 Random

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

Probability and Probability Distributions. Dr. Mohammed Alahmed

Probability and Probability Distributions. Dr. Mohammed Alahmed Probability and Probability Distributions 1 Probability and Probability Distributions Usually we want to do more with data than just describing them! We might want to test certain specific inferences about

More information

TMA 4275 Lifetime Analysis June 2004 Solution

TMA 4275 Lifetime Analysis June 2004 Solution TMA 4275 Lifetime Analysis June 2004 Solution Problem 1 a) Observation of the outcome is censored, if the time of the outcome is not known exactly and only the last time when it was observed being intact,

More information

Machine Learning. Module 3-4: Regression and Survival Analysis Day 2, Asst. Prof. Dr. Santitham Prom-on

Machine Learning. Module 3-4: Regression and Survival Analysis Day 2, Asst. Prof. Dr. Santitham Prom-on Machine Learning Module 3-4: Regression and Survival Analysis Day 2, 9.00 16.00 Asst. Prof. Dr. Santitham Prom-on Department of Computer Engineering, Faculty of Engineering King Mongkut s University of

More information

Nonparametric Bayes Estimator of Survival Function for Right-Censoring and Left-Truncation Data

Nonparametric Bayes Estimator of Survival Function for Right-Censoring and Left-Truncation Data Nonparametric Bayes Estimator of Survival Function for Right-Censoring and Left-Truncation Data Mai Zhou and Julia Luan Department of Statistics University of Kentucky Lexington, KY 40506-0027, U.S.A.

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 February 3, 2015 21-1 Time matching/risk set sampling/incidence density sampling/nested design 21-2 21-3

More information

STATISTICS SYLLABUS UNIT I

STATISTICS SYLLABUS UNIT I STATISTICS SYLLABUS UNIT I (Probability Theory) Definition Classical and axiomatic approaches.laws of total and compound probability, conditional probability, Bayes Theorem. Random variable and its distribution

More information

ADVANCED STATISTICAL ANALYSIS OF EPIDEMIOLOGICAL STUDIES. Cox s regression analysis Time dependent explanatory variables

ADVANCED STATISTICAL ANALYSIS OF EPIDEMIOLOGICAL STUDIES. Cox s regression analysis Time dependent explanatory variables ADVANCED STATISTICAL ANALYSIS OF EPIDEMIOLOGICAL STUDIES Cox s regression analysis Time dependent explanatory variables Henrik Ravn Bandim Health Project, Statens Serum Institut 4 November 2011 1 / 53

More information

Smooth nonparametric estimation of a quantile function under right censoring using beta kernels

Smooth nonparametric estimation of a quantile function under right censoring using beta kernels Smooth nonparametric estimation of a quantile function under right censoring using beta kernels Chanseok Park 1 Department of Mathematical Sciences, Clemson University, Clemson, SC 29634 Short Title: Smooth

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

Economics 508 Lecture 22 Duration Models

Economics 508 Lecture 22 Duration Models University of Illinois Fall 2012 Department of Economics Roger Koenker Economics 508 Lecture 22 Duration Models There is considerable interest, especially among labor-economists in models of duration.

More information

Ph.D. course: Regression models. Introduction. 19 April 2012

Ph.D. course: Regression models. Introduction. 19 April 2012 Ph.D. course: Regression models Introduction PKA & LTS Sect. 1.1, 1.2, 1.4 19 April 2012 www.biostat.ku.dk/~pka/regrmodels12 Per Kragh Andersen 1 Regression models The distribution of one outcome variable

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

Conditional independence of blocked ordered data

Conditional independence of blocked ordered data Conditional independence of blocked ordered data G. Iliopoulos 1 and N. Balakrishnan 2 Abstract In this paper, we prove that blocks of ordered data formed by some conditioning events are mutually independent.

More information

STA 216: GENERALIZED LINEAR MODELS. Lecture 1. Review and Introduction. Much of statistics is based on the assumption that random

STA 216: GENERALIZED LINEAR MODELS. Lecture 1. Review and Introduction. Much of statistics is based on the assumption that random STA 216: GENERALIZED LINEAR MODELS Lecture 1. Review and Introduction Much of statistics is based on the assumption that random variables are continuous & normally distributed. Normal linear regression

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

Parameter Estimation

Parameter Estimation Parameter Estimation Chapters 13-15 Stat 477 - Loss Models Chapters 13-15 (Stat 477) Parameter Estimation Brian Hartman - BYU 1 / 23 Methods for parameter estimation Methods for parameter estimation Methods

More information

1 The problem of survival analysis

1 The problem of survival analysis 1 The problem of survival analysis Survival analysis concerns analyzing the time to the occurrence of an event. For instance, we have a dataset in which the times are 1, 5, 9, 20, and 22. Perhaps those

More information

Pairwise rank based likelihood for estimating the relationship between two homogeneous populations and their mixture proportion

Pairwise rank based likelihood for estimating the relationship between two homogeneous populations and their mixture proportion Pairwise rank based likelihood for estimating the relationship between two homogeneous populations and their mixture proportion Glenn Heller and Jing Qin Department of Epidemiology and Biostatistics Memorial

More information

Goodness-of-fit tests for randomly censored Weibull distributions with estimated parameters

Goodness-of-fit tests for randomly censored Weibull distributions with estimated parameters Communications for Statistical Applications and Methods 2017, Vol. 24, No. 5, 519 531 https://doi.org/10.5351/csam.2017.24.5.519 Print ISSN 2287-7843 / Online ISSN 2383-4757 Goodness-of-fit tests for randomly

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

Concepts and Tests for Trend in Recurrent Event Processes

Concepts and Tests for Trend in Recurrent Event Processes JIRSS (2013) Vol. 12, No. 1, pp 35-69 Concepts and Tests for Trend in Recurrent Event Processes R. J. Cook, J. F. Lawless Department of Statistics and Actuarial Science, University of Waterloo, Ontario,

More information

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

Ph.D. course: Regression models. Regression models. Explanatory variables. Example 1.1: Body mass index and vitamin D status

Ph.D. course: Regression models. Regression models. Explanatory variables. Example 1.1: Body mass index and vitamin D status Ph.D. course: Regression models Introduction PKA & LTS Sect. 1.1, 1.2, 1.4 25 April 2013 www.biostat.ku.dk/~pka/regrmodels13 Per Kragh Andersen Regression models The distribution of one outcome variable

More information

5 Introduction to the Theory of Order Statistics and Rank Statistics

5 Introduction to the Theory of Order Statistics and Rank Statistics 5 Introduction to the Theory of Order Statistics and Rank Statistics This section will contain a summary of important definitions and theorems that will be useful for understanding the theory of order

More information

Review of Multivariate Survival Data

Review of Multivariate Survival Data Review of Multivariate Survival Data Guadalupe Gómez (1), M.Luz Calle (2), Anna Espinal (3) and Carles Serrat (1) (1) Universitat Politècnica de Catalunya (2) Universitat de Vic (3) Universitat Autònoma

More information

PhD course: Statistical evaluation of diagnostic and predictive models

PhD course: Statistical evaluation of diagnostic and predictive models PhD course: Statistical evaluation of diagnostic and predictive models Tianxi Cai (Harvard University, Boston) Paul Blanche (University of Copenhagen) Thomas Alexander Gerds (University of Copenhagen)

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

CIMAT Taller de Modelos de Capture y Recaptura Known Fate Survival Analysis

CIMAT Taller de Modelos de Capture y Recaptura Known Fate Survival Analysis CIMAT Taller de Modelos de Capture y Recaptura 2010 Known Fate urvival Analysis B D BALANCE MODEL implest population model N = λ t+ 1 N t Deeper understanding of dynamics can be gained by identifying variation

More information

Generalized Linear. Mixed Models. Methods and Applications. Modern Concepts, Walter W. Stroup. Texts in Statistical Science.

Generalized Linear. Mixed Models. Methods and Applications. Modern Concepts, Walter W. Stroup. Texts in Statistical Science. Texts in Statistical Science Generalized Linear Mixed Models Modern Concepts, Methods and Applications Walter W. Stroup CRC Press Taylor & Francis Croup Boca Raton London New York CRC Press is an imprint

More information

Hacettepe Journal of Mathematics and Statistics Volume 45 (5) (2016), Abstract

Hacettepe Journal of Mathematics and Statistics Volume 45 (5) (2016), Abstract Hacettepe Journal of Mathematics and Statistics Volume 45 (5) (2016), 1605 1620 Comparing of some estimation methods for parameters of the Marshall-Olkin generalized exponential distribution under progressive

More information

STAT 526 Spring Final Exam. Thursday May 5, 2011

STAT 526 Spring Final Exam. Thursday May 5, 2011 STAT 526 Spring 2011 Final Exam Thursday May 5, 2011 Time: 2 hours Name (please print): Show all your work and calculations. Partial credit will be given for work that is partially correct. Points will

More information

Copula Regression RAHUL A. PARSA DRAKE UNIVERSITY & STUART A. KLUGMAN SOCIETY OF ACTUARIES CASUALTY ACTUARIAL SOCIETY MAY 18,2011

Copula Regression RAHUL A. PARSA DRAKE UNIVERSITY & STUART A. KLUGMAN SOCIETY OF ACTUARIES CASUALTY ACTUARIAL SOCIETY MAY 18,2011 Copula Regression RAHUL A. PARSA DRAKE UNIVERSITY & STUART A. KLUGMAN SOCIETY OF ACTUARIES CASUALTY ACTUARIAL SOCIETY MAY 18,2011 Outline Ordinary Least Squares (OLS) Regression Generalized Linear Models

More information