Computing optimal sampling designs for two-stage studies



Stata Technical Bulletin

17


Koopman’s method

Let X and Y be two independent binomial variates based on sample sizes m and n and parameters p1 and p2, respectively.
Let
θ = pι∕p2∙ Koopman (1984) proposed a method for constructing confidence intervals for θ based on a chi-squared test.
oopman’s method has been widely used in medical research for evaluating drug efficacy and treatment effects.

Assume we test for H0 : θ = Θq against Ha : O ≠ 0o∙ For this problem there is no uniformly most powerful test as extreme
values may occur in the sample, but a chi-squared test seems a reasonable choice. The test statistic
Ug0 is then given by

ττ t ч {x - τnp1}2 (y - np2}2

θo x,y mp1 (I-Pi) np2(1 - p2)

where pɪ and p2 are the maximum likelihood estimates under the restriction θ = 0o, It can be proved that

ʌ _ θ0(m + y) + x + n- [{0o(m + ?/) + x + n}2 - ⅛θ0(m + n)(x + y)]1/2

2(m + n)

andp2 = pi∕0o.

For 0 = 1, the statistic Ug0(x,y) is the traditional Pearson chi-square. Rearranging Ug0(x,y) results in

rr ( ʌ (ж-Wi)2 ʃ m(0θ-Pι)l

This shows that under H0, Ug0{x, y) has asymptotically for m —> ∞ and n —> ∞, a chi-squared distribution with 1 degree
of freedom independent of
Θq (Bishop et al. 1977). Hence, an approximate 1 — α two-sided confidence region for θ is given by

{Uθo(x,y) < X1,1-α}

where χ2 1-α is the 1 — α fractile of the chi-squared distribution with 1 degree of freedom. Since U is a convex function of θ,
this is an asymmetric interval (0∕,0∙u), where

l(x,y) = UθJx,y} = χ2 ι-o.

and

θι < θu

As Ugl (ж, ye reduces to the usual chi-squared when 0 = 1, this interval will always agree with the chi-squared test.

Because there is no explicit expression for the inverse function of U, the values of 0/ and θu have to be solved by numerical
procedures. The main concern of the command koopman is to obtain 0/ and
θu by using repeated bisection as suggested by
Koopman (1984).

Syntax

koopman vαrxevent var.group [weihht [if ixp [in range [, level (#) ]

koopmani #x #m #y #n [, level (#) ]

koopman allows fweights.

Description

koopman computes confidence intervals for the ratio of two binomial proportions based on two independent binomially
distributed random variables using Koopman’s method. Point estimates and confidence intervals for the odds ratio are calculated.
ivent-var contains a one if the observation represents an event and 0 otherwise. group~var indicates the group to which each
observation belongs. The variable must have only two values. Observations with missing values are not used.

koopmani is the immediate form of koopman.



More intriguing information

1. The name is absent
2. Regional science policy and the growth of knowledge megacentres in bioscience clusters
3. Manufacturing Earnings and Cycles: New Evidence
4. Learning-by-Exporting? Firm-Level Evidence for UK Manufacturing and Services Sectors
5. Better policy analysis with better data. Constructing a Social Accounting Matrix from the European System of National Accounts.
6. On the Desirability of Taxing Charitable Contributions
7. Structural Breakpoints in Volatility in International Markets
8. Pupils’ attitudes towards art teaching in primary school: an evaluation tool
9. Road pricing and (re)location decisions households
10. On the Existence of the Moments of the Asymptotic Trace Statistic