CSU Hayward

Statistics Department

Cast of "Characters"

Symbols Used in "Elementary Uses of the Gibbs Sampler: Applications to Medical Screening Tests"
STATS #27, Winter 2000, by Suess, Fraser, and Trumbo


 :
Character Name Definition or Usage
A   Observed count, #(T=1)
a, b alpha, beta   Parameters of beta prior distribution.
Transition probabilities of a 2-state Markov Chain
B   Observed count, #(T=0)
D   Random variable: 1 if virus present, 0 otherwise
d delta Predictive value of negative test: P(D=0|T=0);  d* = 1 – d
h eta Sensitivity: P(T=1|D=1);  h* = 1 – h
g gamma Predictive value of positive test: P(D=1|T=1):  g* = 1 – g
m   Simulation step number
M1   Number of simulation steps until burn in
M2   Total number of simulation steps
N   Sample size
p   Estimate of p
p pi Prevalence: P(D=1);  p* = 1 – p
P Pi Random variable for prevalence in Gibbs Sampler
t   Estimate of t
T   Random variable: 1 if positive test, 0 otherwise
t tau Probability of positive test: P(T=1)
q theta Specificity: P(T=0|D=0);  q* = 1 – q
X   Observed or simulated count, #(T=1, D=1)
Y   Observed or simulated count, #(T=0, D=1)