Paper: Biostatistics
Principal investigator: Dr.Bhaswati Ganguli, Department of Statistics, University of Calcutta
Paper co-ordinator: Dr.Sugata SenRoy,Department of Statistics, University of Calcutta
Content writer: Dr.Atanu Bhattacharjee, Division of Clinical Research and Biostatistics, Malabar Cancer Centre Content reviewer: Dr.Indranil Mukhopadhyay,Indian Statistical
Institute, Kolkata
* Each pair has two members.
* Each sampled are independently provide the value of the matching covariate Z.
* The members within each pair are conditionally independent.
* The sample of matched-pairs be considered to consist ofN independent samples(strata).
* Each pair has two members.
* Each sampled are independently provide the value of the matching covariate Z.
* The members within each pair are conditionally independent.
* The sample of matched-pairs be considered to consist ofN independent samples(strata).
* Each pair has two members.
* Each sampled are independently provide the value of the matching covariate Z.
* The members within each pair are conditionally independent.
* The sample of matched-pairs be considered to consist ofN independent samples(strata).
* Each pair has two members.
* Each sampled are independently provide the value of the matching covariate Z.
* The members within each pair are conditionally independent.
* The sample of matched-pairs be considered to consist ofN independent samples(strata).
* A2×2 table is obtained from the measures of each strata.
* SupposeK represents the total number of strata and j= 1, ...., K.
* The strata may be formed through categories of a single covariate ( may be sex)
* The strata wise observation can be formed through table
* A2×2 table is obtained from the measures of each strata.
* SupposeK represents the total number of strata and j= 1, ...., K.
* The strata may be formed through categories of a single covariate ( may be sex)
* The strata wise observation can be formed through table
* A2×2 table is obtained from the measures of each strata.
* SupposeK represents the total number of strata and j= 1, ...., K.
* The strata may be formed through categories of a single covariate ( may be sex)
* The strata wise observation can be formed through table
* A2×2 table is obtained from the measures of each strata.
* SupposeK represents the total number of strata and j= 1, ...., K.
* The strata may be formed through categories of a single covariate ( may be sex)
* The strata wise observation can be formed through table
Title1:- Jth stratum Frequency Table Response Group1 Grop2 Response(+) aj bj m1j
Response(-) cj dj m2j n1j n2j Nj Title2:- Jth stratum Probability Table
Response Group1 Grop2 Response(+) π1j π2j
Response(-) 1-π1j 1-π2j
Total 1 1
* Within thejth stratum any of the measures of group-response association described previously may be computed:
* The risk difference RDˆ j =p1j−p2j,
* Relative Risk RRˆ j = pp1j
2j = abjn2j
jn1j
* The odds ratioORˆj = 1−p1jp
1j/[1−p2jp
2j] = abjdj
jcj.
* Within thejth stratum any of the measures of group-response association described previously may be computed:
* The risk difference RDˆ j =p1j−p2j,
* Relative Risk RRˆ j = pp1j
2j = abjn2j
jn1j
* The odds ratioORˆj = 1−p1jp
1j/[1−p2jp
2j] = abjdj
jcj.
* Within thejth stratum any of the measures of group-response association described previously may be computed:
* The risk difference RDˆ j =p1j−p2j,
* Relative Risk RRˆ j = pp1j
2j = abjn2j
jn1j
* The odds ratioORˆj = 1−p1jp
1j/[1−p2jp
2j] = abjdj
jcj.
* Within thejth stratum any of the measures of group-response association described previously may be computed:
* The risk difference RDˆ j =p1j−p2j,
* Relative Risk RRˆ j = pp1j
2j = abjn2j
jn1j
* The odds ratioORˆj = 1−p1jp
1j/[1−p2jp
2j] = abjdj
jcj.
* Test of the null hypothesisH0 :π1j =π2j(ORj= 1) for all j.
* The alternative hypothesis that the probabilities within strata differ such that there is a common odds ratio i.e. H1: (ORj =OR̸= 1)∀j= 1,2, ....K.
* Test of the null hypothesisH0 :π1j =π2j(ORj= 1) for all j.
* The alternative hypothesis that the probabilities within strata differ such that there is a common odds ratio i.e. H1: (ORj =OR̸= 1)∀j= 1,2, ....K.
* The expected frequency for the index cell,E(aj)
* Ej =E(aj) = n1jNm1j
j
* Variance ofaj underH0 is
* Vcj = m1jNm2 2jn1jn2j
j(Nj−1) For each statum , under H0 asymptotically for largeNj and for fixed K
* The expected frequency for the index cell,E(aj)
* Ej =E(aj) = n1jNm1j
j
* Variance ofaj underH0 is
* Vcj = m1jNm2 2jn1jn2j
j(Nj−1) For each statum , under H0 asymptotically for largeNj and for fixed K
* The expected frequency for the index cell,E(aj)
* Ej =E(aj) = n1jNm1j
j
* Variance ofaj underH0 is
* Vcj = m1jNm2 2jn1jn2j
j(Nj−1) For each statum , under H0 asymptotically for largeNj and for fixed K
* The expected frequency for the index cell,E(aj)
* Ej =E(aj) = n1jNm1j
j
* Variance ofaj underH0 is
* Vcj = m1jNm2 2jn1jn2j
j(Nj−1) For each statum , under H0 asymptotically for largeNj and for fixed K
* The termaj−Ej ≈N(0, Vcj).
* ∑
j(aj−Ej)≈N[0,∑
jVcj]
* The stratified-adjusted Mantel-Haenszel is XC(M H)2 = [
∑
j(aj−Ej)]2
∑
jVcj = [a+V−E+]2
c+
a+ =∑K
j=iaj, E+=∑
jEj, and Vc+=∑
jVcj.
* It is noted that(a+−E+) is the sum of asymptotically normally distributed then XC(M H)2 ≈χ2 on 1 d.f.
* The termaj−Ej ≈N(0, Vcj).
* ∑
j(aj−Ej)≈N[0,∑
jVcj]
* The stratified-adjusted Mantel-Haenszel is XC(M H)2 = [
∑
j(aj−Ej)]2
∑
jVcj = [a+V−E+]2
c+
a+ =∑K
j=iaj, E+=∑
jEj, and Vc+=∑
jVcj.
* It is noted that(a+−E+) is the sum of asymptotically normally distributed then XC(M H)2 ≈χ2 on 1 d.f.
* The termaj−Ej ≈N(0, Vcj).
* ∑
j(aj−Ej)≈N[0,∑
jVcj]
* The stratified-adjusted Mantel-Haenszel is XC(M H)2 = [
∑
j(aj−Ej)]2
∑
jVcj = [a+V−E+]2
c+
a+ =∑K
j=iaj, E+=∑
jEj, and Vc+=∑
jVcj.
* It is noted that(a+−E+) is the sum of asymptotically normally distributed then XC(M H)2 ≈χ2 on 1 d.f.
* The termaj−Ej ≈N(0, Vcj).
* ∑
j(aj−Ej)≈N[0,∑
jVcj]
* The stratified-adjusted Mantel-Haenszel is XC(M H)2 = [
∑
j(aj−Ej)]2
∑
jVcj = [a+V−E+]2
c+
a+ =∑K
j=iaj, E+=∑
jEj, and Vc+=∑
jVcj.
* It is noted that(a+−E+) is the sum of asymptotically normally distributed then XC(M H)2 ≈χ2 on 1 d.f.
* The termaj−Ej ≈N(0, Vcj).
* ∑
j(aj−Ej)≈N[0,∑
jVcj]
* The stratified-adjusted Mantel-Haenszel is XC(M H)2 = [
∑
j(aj−Ej)]2
∑
jVcj = [a+V−E+]2
c+
a+ =∑K
j=iaj, E+=∑
jEj, and Vc+=∑
jVcj.
* It is noted that(a+−E+) is the sum of asymptotically normally distributed then XC(M H)2 ≈χ2 on 1 d.f.
* The termaj−Ej ≈N(0, Vcj).
* ∑
j(aj−Ej)≈N[0,∑
jVcj]
* The stratified-adjusted Mantel-Haenszel is XC(M H)2 = [
∑
j(aj−Ej)]2
∑
jVcj = [a+V−E+]2
c+
a+ =∑K
j=iaj, E+=∑
jEj, and Vc+=∑
jVcj.
* It is noted that(a+−E+) is the sum of asymptotically normally distributed then XC(M H)2 ≈χ2 on 1 d.f.
1:M matching
Note there are2(M+ 1) possible outcomes 2 ofn1 < mare concordant and hence excluded. Divide the remaining2M into M sets of 2 as Title1:- 2x2 contigency table
Exposed Not-exposed
Cases Exposed 1 0
Control m-1 M-m+1
m M-m+1
1:M matching
Pn= (Mm−1)p1pm0−1(1−p0)M−m+1 (1) Title2:- 2x2 contigency table
Control Exposed Control Not-exposed
Cases Exposed 0 1
Cases Not-exposed m M-m
m M-m+1
form= 1,2, ....M
1:M matching
Pn= (Mm)pm0 (1−p0)M−m(1−p1) (2) Conditional probability of case exposed /M exposed
πm=P r[Case exposed|mExposed] (3) πm= (Mm−1)p1pm0 −1(1−p0)M−m+1
(Mm−1)p1pm0−1(1−p0)M−m+1+ (Mm)pm0 (1−p0)M−m(1−p1) (4)
πm = (1−p0)p1
p1(1−p0) +M−mm+1p0(1−p1) (5) πm = mψ
mψ+M −m+ 1 (6) Let,ni,jm= i cases and jm controls exposed, i=0,1.Then,
n1,m−1 ∼Bin(n1,m−1+n0,m, πm), m= 1(1)M LE =
∏m
( mψ
mψ+M−m+ 1)n1,m−1( M−m+ 1
mψ+M−m+ 1)n0,m
1:M matching Also,
E(n1,m−1|n1,m−1+n0,m) = (n1,m−1+n0,m)mψ
mψ+M−m+ 1 (8) V ar(n1,m−1|n1,m−1+n0,m) = (n1,m−1+n0,m)mψ(M−m+ 1)
(mψ+M −m+ 1)2 (9)
1:M matching NowL=log L=∑m
m=1[n1,m−1{log(mψ)−log(mψ+M −m+ 1)}+ n0,m{log(M −m+ 1)−log(mψ+M−m+ 1)}]
1:M matching
δL δψ =
∑M m=1
1 ψ =
∑m m=1
m
mψ+M−m+ 1{n0,m+n1,m−1} (10)
∑M m=1
n1,m−1 =
∑M m=1
(n0,m+n1,m−1)mψˆ
mψˆ+M−m+ 1 (11)
1:M matching Thus,
ψˆM H =
∑M
m=1(M−m+ 1)n1,m−1
∑M
m=1mn0,m
(12) To testH0 :ψ= 1, we can see
χ2= [1∑M
m=1(n1,m−1−(n1,m−1M+1+n0,m−1)m)1− 12]2
1 (M+1)2
∑M
m=1(n1,m−1+n0,m−1)m(M −m+ 1) (13)