Dr Chris Drovandi, statistical science lecturer at QUT, discusses a general framework for likelihood-free Bayesian inference problems called Bayesian Indirect Likelihood (BIL). There is a focus on some specific instances of the BIL framework that use in some way a parametric auxiliary model, which is an alternative model that possesses a tractable likelihood function. These methods are referred to as parametric Bayesian Indirect Inference (pBII) methods.
One class of pBII methods uses the score or parameter of the auxiliary model to form summary statistics for ABC. A different class (called parametric BIL, pBIL) uses the likelihood (either at the full-data or summary statistic level) of the auxiliary model as a replacement to the likelihood of the true model.
The theoretical aspects of pBIL are explored and contrasted against the ABC approach that uses summary statistics formed from the auxiliary model. The theoretical results and the performance of the methods will be demonstrated on examples of varying complexity, including estimating the parameters of a stochastic process for macroparasite population evolution.
This is joint work with Prof. Tony Pettitt, Dr Anthony Lee and Leah South.
Identify Rules that Predict Patient’s Heart Disease - An Application of Decis...
Dr Chris Drovandi (QUT) - Bayesian Indirect Inference Using a Parametric Auxiliary Model
1. Bayesian Indirect Likelihood pBII Methods Examples References
Bayesian Indirect Inference using a Parametric
Auxiliary Model
Dr Chris Drovandi
Queensland University of Technology, Australia
c.drovandi@qut.edu.au
Collaborators: Tony Pettitt and Anthony Lee
July 25, 2014
Chris Drovandi QUT Seminar 2014
2. Bayesian Indirect Likelihood pBII Methods Examples References
Bayesian Indirect Likelihood
Bayesian inference is based on posterior distribution
p(θ|Ý ) ∝ p(Ý |θ)p(θ).
Thus Bayesian inference requires the likelihood function
p(Ý |θ)
Many models have an intractable likelihood function
Replace with some alternative tractable ‘likelihood’
˜p(θ|Ý ) ∝ ˜p(Ý |θ)p(θ).
Referred to as Bayesian Indirect Likelihood (BIL)
Chris Drovandi QUT Seminar 2014
3. Bayesian Indirect Likelihood pBII Methods Examples References
BIL Methods
Composite Likelihood (e.g. Pauli et al 2011)
Emulation (e.g. Gaussian process, Wilkinson 2014)
parametric BIL (uses the likelihood of an alternative
parametric model, e.g. Gallant and McCullagh 2009)
non-parametric BIL (traditional ABC method recovered, e.g.
Beaumont et al 2002)
Chris Drovandi QUT Seminar 2014
4. Bayesian Indirect Likelihood pBII Methods Examples References
pBII methods
pBII - Bayesian Indirect Inference methods that uses a
‘p’arametric auxiliary model in some way
Requires parametric model with parameter φ with tractable
likelihood, pA(y|φ)
Does not have to be an obvious connection between θ and φ
Assume throughout that dim(φ) ≥ dim(θ)
Two classes of pBII methods:
Use auxiliary model to form summary statistics for ABC (ABC
II)
Use auxiliary model to form replacement likelihood, pBIL
Chris Drovandi QUT Seminar 2014
5. Bayesian Indirect Likelihood pBII Methods Examples References
ABC II (Intro to ABC)
ABC assumes that simulation from model is straightforward,
Ü ∼ p(Ý |θ)
Compares observed and simulated data on the basis of
summary statistics, s(·)
Based on the following target distribution
pǫ(θ|Ý ) ∝ p(θ)pǫ(Ý |θ),
where ǫ is ABC tolerance and
pǫ(Ý |θ) =
Ü
p(Ü|θ)Kǫ(||s(Ý ) − s(Ü )||)dx,
where K is a kernel weighting function. This can be estimated
unbiasedly and this is enough (Andrieu and Roberts 2009)
If s(·) is sufficient and ǫ → 0 then pǫ(θ|Ý ) ≡ p(θ|Ý ) (does not
happen in practice)
Thus two sources of error
Chris Drovandi QUT Seminar 2014
6. Bayesian Indirect Likelihood pBII Methods Examples References
ABC II (Intro to ABC Cont...)
ABC has connections with kernel density estimation (Blum,
2009)
Choice of summary statistics involve trade-off between
dimensionality and information loss
Non-parametric aspect wants summary as low-dimensional as
possible
But decreasing dimension means information loss
Choice of summary statistics most crucial to ABC
approximation
Chris Drovandi QUT Seminar 2014
7. Bayesian Indirect Likelihood pBII Methods Examples References
ABC II Methods
In some applications can propose an alternative parametric
auxiliary model
Use auxiliary model to form summary statistic (ABC II)
Auxiliary parameter estimate as summary statistic (ABC IP
and ABC IL)
Score of auxiliary model as summary statistic (ABC IS)
Chris Drovandi QUT Seminar 2014
8. Bayesian Indirect Likelihood pBII Methods Examples References
ABC IP (Drovandi et al 2011)
Parameter estimates of auxiliary model are summary statistics
Simulate data from generative model Ü ∼ p(·|θ)
Estimate auxiliary parameter
φ(θ, Ü) = arg max
φ
pA(Ü|φ).
Compare with φ(Ý )
ρ(s(Ü ), s(Ý )) = (φ(Ü) − φ(Ý ))T Á (φ(Ý ))(φ(Ü ) − φ(Ý )).
Efficient weighting of summary statistics using observed Fisher
information Á (φ(Ý ))
Chris Drovandi QUT Seminar 2014
9. Bayesian Indirect Likelihood pBII Methods Examples References
ABC IL (Gleim and Pigorsch 2013)
Same as ABC IP but uses likelihood in the discrepancy
function
Parameter estimates of auxiliary model are summary statistics
Simulate data from generative model Ü ∼ p(·|θ)
Estimate auxiliary parameter
φ(θ, Ü) = arg max
φ
pA(Ü|φ).
Compare with φ(Ý )
ρ(s(Ü ), s(Ý )) = log pA(Ý |φ(Ý )) − log pA(Ý |φ(Ü)).
Chris Drovandi QUT Seminar 2014
10. Bayesian Indirect Likelihood pBII Methods Examples References
ABC IS (Gleim and Pigorsch 2013)
Uses scores of auxiliary model as summary statistics
Ë (Ý , φ) =
∂ log pA(Ý |φ)
∂φ1
, · · · ,
∂ log pA(Ý |φ)
∂φdim(φ)
T
,
Simulate data from generative model Ü ∼ p(·|θ)
Evaluate scores of auxiliary model based on simulated data
and φ(Ý ), Ë (Ü, φ(Ý ))
Note that Ë (Ý , φ(Ý )) = 0. Uses following discrepancy
function
ρ(s(Ü ), s(Ý )) = Ë (Ü, φ(Ý ))T Á (φ(Ý ))−1Ë (Ü , φ(Ý )).
Very fast when scores are analytic (no fitting of auxiliary
model)
Chris Drovandi QUT Seminar 2014
11. Bayesian Indirect Likelihood pBII Methods Examples References
ABC II Assumptions
Assumption (ABC IP Assumptions)
The estimator of the auxiliary parameter, φ(θ, Ü ), is unique for all
θ with positive prior support.
Assumption (ABC IL Assumptions)
The auxiliary likelihood evaluated at the auxiliary estimate,
pA(Ý |φ(Ü , θ)), is unique for all θ with positive prior support.
Assumption (ABC IS Assumptions)
The MLE of the auxiliary model fitted to the observed data, φ(Ý ),
is an interior point of the parameter space of φ and Â(φ(Ý )) is
positive definite. The log-likelihood of the auxiliary model,
log pA(·|φ), is differentiable and the score, ËA(Ü, φ(Ý )), is unique
for any Ü that may be drawn according to any θ that has positive
prior support.
Chris Drovandi QUT Seminar 2014
12. Bayesian Indirect Likelihood pBII Methods Examples References
ABC II vs Traditional ABC for Summary Statistics
Advantages
Can check if auxiliary model fits data well
Natural ABC discrepancy functions
Control dimensionality of summary statistics via parsimonious
auxiliary model (e.g. compare auxiliary models via AIC)
Disadvantages
ABC II can be expensive
Restricted to applications where auxiliary model can be
proposed
Chris Drovandi QUT Seminar 2014
13. Bayesian Indirect Likelihood pBII Methods Examples References
pdBIL (Gallant and McCullogh 2009, Reeves and Pettitt
2005)
Replaces true likelihood with auxiliary likelihood (full data
level)
Simulate n independent datasets from generative model
Ü1:n
iid
∼ p(·|θ)
Estimate auxiliary parameter
φ(θ, Ü1:n) = arg max
φ
pA(Ü1:n|φ).
Provides estimate of mapping φ(θ)
Evaluate auxiliary likelihood pA(Ý |φ(θ, Ü1:n))
Not ABC. No summary statistics or ABC tolerance
Theoretically behaves very different to ABC II
Chris Drovandi QUT Seminar 2014
14. Bayesian Indirect Likelihood pBII Methods Examples References
pdBIL (Cont...)
Ultimate target of pdBIL method
pA(θ|Ý ) ∝ pA(Ý |φ(θ))p(θ).
Unfortunately binding function unknown, estimate via n iid
simulations from generative model, φθ,n = φ(θ, Ü 1:n)
Target distribution of the method becomes
pA,n(θ|Ý ) ∝ pA,n(Ý |θ)p(θ),
where
pA,n(Ý |θ) =
Ü1:n
pA(Ý |φθ,n)p(Ü1:n|θ)dÜ 1:n
pA,n(Ý |θ) can be estimated unbiasedly, which is enough
(Andrieu and Roberts (2009))
Chris Drovandi QUT Seminar 2014
15. Bayesian Indirect Likelihood pBII Methods Examples References
pdBIL (Cont...)
Comparing pA,n(θ|Ý ) with pA(θ|Ý )
Under suitable conditions pA,n(θ|Ý ) → pA(θ|Ý ) as n → ∞
In reality must choose finite n
Unfortunately E[pA(Ý |φθ,n)] = pA(Ý |φ(θ)) thus
pA,n(θ|Ý ) = pA(θ|Ý ) for finite n
Empirical evidence suggests loss of precision as n decreases
Can anticipate better approximations for pBIL for n large
Chris Drovandi QUT Seminar 2014
16. Bayesian Indirect Likelihood pBII Methods Examples References
pdBIL (Cont...)
Comparing pA(θ|Ý ) with p(θ|Ý )
Under suitable conditions pdBIL is exact as n → ∞ if
generative model special case of auxiliary model
Suggests in this context auxiliary model needs to be flexible to
mimic behaviour of generative model
Empirical evidence suggests for good approximation auxiliary
likelihood must do well in non-negligible posterior regions
Chris Drovandi QUT Seminar 2014
17. Bayesian Indirect Likelihood pBII Methods Examples References
Synthetic Likeihood as a psBIL Method
Reduces data to set of summary statistics. Target distribution:
p(θ|s(Ý )) ∝ p(s(Ý )|θ)p(θ).
But p(s(Ý )|θ) is unknown
Synthetic likelihood approach Wood (2010) assumes a
parametric model, pA(s(Ý )|φ(θ)). Thus a pBIL method.
Wood takes pA to be multivariate normal N(µ(θ), Σ(θ))
For some θ, simulate n iid replicates,
Ü1:n = (s(Ü 1), . . . , s(Ü n)), iid sample from p(s(Ý )|θ). Fit
auxiliary model to this dataset:
µ(Ü 1:n, θ) =
1
n
n
i=1
s(Üi ),
Σ(Ü 1:n, θ) =
1
n
n
i=1
(s(Üi ) − µ(Ü1:n, θ))(s(Ü i ) − µ(Ü1:n, θ))T
,
Chris Drovandi QUT Seminar 2014
18. Bayesian Indirect Likelihood pBII Methods Examples References
ABC as BIL with Non-Parametric Auxiliary Model (npBIL)
ABC is recovered as a BIL method by selecting a
non-parametric auxiliary likelihood
Full data (npdBIL): Choose φ(θ, Ü 1:n) = Ü1:n
pA(Ý |φ(θ, Ü 1:n)) = pA(Ý |Ü 1:n) =
1
n
n
i=1
Kǫ(ρ(Ý , Ü i )),
Data reduced to summary statistic (npsBIL):
1
n
n
i=1
Kǫ(ρ(×(Ý ), ×(Ü i ))).
ABC II is a special npsBIL method where summary statistics
come from a parametric auxiliary model
Chris Drovandi QUT Seminar 2014
20. Bayesian Indirect Likelihood pBII Methods Examples References
BIL
pBIL npBIL=ABC
npdBIL npsBIL
ABC II
ABC IP ABC IL ABC IS
pdBIL psBIL
Key:
BIL - ’B’ayesian ’I’ndirect ’L’ikelihood
pBIL - ’p’arametric BIL
pdBIL - pBIL based on full ’d’ata
psBIL - pBIL based on ’s’ummary statistic
ABC - ’a’pproximate ’B’ayesian ’c’omputation
npBIL - ’n’on-’p’arametric BIL (equivalent to ABC)
npdBIL - npBIL based on full ’d’ata
npsBIL - npBIL based on ’s’ummary statistic
ABC II - ABC ’I’ndirect ’I’nference
ABC IP - ABC ’I’ndirect ’P’arameter
ABC IL - ABC ’I’ndirect ’L’ikelihood
ABC IS - ABC ’I’ndirect ’S’core
Figure: BIL framework
Chris Drovandi QUT Seminar 2014
21. Bayesian Indirect Likelihood pBII Methods Examples References
Comparison of pdBIL and ABC II
Theoretical Comarison
pdBIL exact when true model contained within auxiliary model
(as n → ∞). Suggests to choose flexible auxiliary model
Even when true model is special case, ABC II statistics not
sufficient in general
When ABC II have sufficient statistics, pdBIL not exact in
general, even when n → ∞
Some Toy Examples Later...
Choice of Auxiliary Model
ABC II requires good summarisation of data (independent of
specification of generative model)
Auxiliary model for pdBIL does not necessarily have to fit data
well (if generative model mis-specified). Requires flexibility
wrt chosen generative model
If generative model (approx) correct and fits data well then
same auxiliary model may be chosen
Chris Drovandi QUT Seminar 2014
22. Bayesian Indirect Likelihood pBII Methods Examples References
Sampling from ABC II Target - MCMC ABC
Algorithm 1 MCMC ABC algorithm of Marjoram et al 2003.
1: Set θ0
2: for i = 1 to T do
3: Draw θ∗ ∼ q(·|θi−1)
4: Simulate x∗ ∼ p(·|θ∗)
5: Compute r = p(θ∗)q(θi−1|θ∗)
p(θi−1)q(θ∗|θi−1)
1(ρ(s(x∗), s(y)) ≤ ǫ)
6: if uniform(0, 1) < r then
7: θi = θ∗
8: else
9: θi = θi−1
10: end if
11: end for
Chris Drovandi QUT Seminar 2014
23. Bayesian Indirect Likelihood pBII Methods Examples References
Sampling from pdBIL Target - MCMC pdBIL
Find increase in acceptance probability with increase in n
Algorithm 1 MCMC BIL algorithm (see also Gallant and McCullogh 2009).
1: Set θ0
2: Simulate x∗
n ∼ p(·|θ0)
3: Compute φ0 = arg maxφ pA(x∗
n|φ)
4: for i = 1 to T do
5: Draw θ∗ ∼ q(·|θi−1)
6: Simulate x∗
n ∼ p(·|θ∗)
7: Compute ˆφ(x∗
n) = arg maxφ pA(x∗
n|φ)
8: Compute r = pA(y| ˆφ(x∗
n))π(θ∗)q(θi−1|θ∗)
pA(y|φi−1)π(θi−1)q(θ∗|θi−1)
9: if uniform(0, 1) < r then
10: θi = θ∗
11: φi = ˆφ(x∗
n)
12: else
13: θi = θi−1
14: φi = φi−1
15: end if
16: end for
Chris Drovandi QUT Seminar 2014
24. Bayesian Indirect Likelihood pBII Methods Examples References
Toy Example 1 (Drovandi and Pettitt 2013)
True model Poisson(λ) and auxiliary model Normal(µ, τ)
ABC II ‘gets lucky’ as ˆµ = ¯y, sufficient for λ
pdBIL as n → ∞ approximates Poisson(λ) likelihood with
Normal(λ, λ) likelihood. Not exact.
28 30 32
0
0.2
0.4
0.6
0.8
n = 1
n = 10
n = ∞
true
20 30 40 50
−800
−600
−400
λ
log−likelihood
true
auxiliary
Acceptance probabilities: n = 1 46%, n = 10 67%, n = 100 72%
and n = 1000 73% for increasing n
Chris Drovandi QUT Seminar 2014
25. Bayesian Indirect Likelihood pBII Methods Examples References
Toy Example 1
Results for pdBIL when auxiliary model mis-specified
Normal(µ, 9) (underdispersed) Normal(µ, 49) (overdispersed)
ABC II will still work by chance
28 30 32
0
0.2
0.4
0.6
0.8
1
true
τ unspecified
τ = 16
τ = 49
29 30 31 32
−330
−320
−310
λ
log−likelihood true
τ unspecified
τ=16
τ=49
Chris Drovandi QUT Seminar 2014
26. Bayesian Indirect Likelihood pBII Methods Examples References
Toy Example 2
True model t-distribution(µ, σ, ν = 1) and auxiliary model
t-distribution(µ, σ, ν)
Here pdBIL exact as n → ∞ since true is special case of
auxiliary
ABC II do not produce sufficient statistics (full set of order
statistics are minimal sufficient)
−1.5 −1 −0.5 0 0.5 1 1.5
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
true
n=1
n=10
n=100
0.5 1 1.5 2 2.5
0
0.5
1
1.5
2
2.5
true
n=1
n=10
n=100
Chris Drovandi QUT Seminar 2014
27. Bayesian Indirect Likelihood pBII Methods Examples References
Toy Example 2 cont
−1 −0.5 0 0.5 1
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
true
ABC IP
ABC IL
ABC IS
0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6
0
0.5
1
1.5
2
2.5
true
ABC IP
ABC IL
ABC IS
Chris Drovandi QUT Seminar 2014
28. Bayesian Indirect Likelihood pBII Methods Examples References
Quantile Distribution Example
Model of Interest: g-and-k quantile distribution
Defined in terms of its quantile function:
Q(z(p); θ) = a + b 1 + c
1 − exp(−gz(p))
1 + exp(−gz(p))
(1 + z(p)2
)k
z(p).
(1)
p - quantile, z(p) - standard normal quantile, θ = (a, b, g, k),
c = 0.8 (see Rayner and MacGillivray 2003).
Numerical likelihood evaluation possible
Simulation easier via inversion method
Data consists of 10000 independent draws with a = 3, b = 1,
g = 2 and k = 0.5
Chris Drovandi QUT Seminar 2014
29. Bayesian Indirect Likelihood pBII Methods Examples References
Quantile Distribution Example (Cont...)
Auxiliary model is a 3-component normal mixture model
Flexible and fits data well
But breaks assumption of ABC IP (parameter estimates not
unique)
0 5 10 15 20
0
0.2
0.4
0.6
0.8
y
density
estimated density
3 component mixture density
Chris Drovandi QUT Seminar 2014
30. Bayesian Indirect Likelihood pBII Methods Examples References
pdBIL Results
2.95 3 3.05
0
10
20
30
n=1
n=10
n=50
posterior
(a) a
0.9 1 1.1
0
5
10
15
(b) b
1.9 2 2.1 2.2
0
5
10
(c) g
0.45 0.5 0.55
0
10
20
30
(d) k
Acc Prob: 2.8% for n = 1, 13.1% for n = 10 and 20.8% for n = 50
Chris Drovandi QUT Seminar 2014
31. Bayesian Indirect Likelihood pBII Methods Examples References
ABC II (no regression adjustment)
2.95 3 3.05
0
10
20
30
posterior
ABC IP
ABC IL
ABC IS
(e) a
0.9 1 1.1
0
5
10
15
(f) b
1.9 2 2.1 2.2
0
5
10
(g) g
0.45 0.5 0.55
0
10
20
30
(h) k
Chris Drovandi QUT Seminar 2014
32. Bayesian Indirect Likelihood pBII Methods Examples References
Compare all (note ABC has regression adjustment)
2.95 3 3.05
0
10
20
30
ABC
ABC IS
ABC IL
ABC IP
pdBIL n=50
posterior
(i) a
0.9 1 1.1
0
5
10
15
(j) b
1.9 2 2.1 2.2
0
5
10
(k) g
0.45 0.5 0.55
0
10
20
30
(l) k
Chris Drovandi QUT Seminar 2014
33. Bayesian Indirect Likelihood pBII Methods Examples References
4 Component Auxiliary Mixture Model
2.95 3 3.05
0
10
20
30
40
ABC
ABC IS
ABC IL
ABC IP
pdBIL n=50
posterior
(m) a
0.9 1 1.1
0
5
10
15
(n) b
1.9 2 2.1 2.2
0
5
10
(o) g
0.45 0.5 0.55
0
10
20
30
(p) k
Chris Drovandi QUT Seminar 2014
34. Bayesian Indirect Likelihood pBII Methods Examples References
Macroparasite Immunity Example
Estimate parameters of a Markov process model explaining
macroparasite population development with host immunity
212 hosts (cats) i = 1, . . . , 212. Each cat injected with li
juvenile Brugia pahangi larvae (approximately 100 or 200).
At time ti host is sacrificed and the number of matures are
recorded
Host assumed to develop an immunity
Three variable problem: M(t) matures, L(t) juveniles, I(t)
immunity.
Only L(0) and M(ti ) is observed for each host
No tractable likelihood
Chris Drovandi QUT Seminar 2014
36. Bayesian Indirect Likelihood pBII Methods Examples References
Trivariate Markov Process of Riley et al (2003)
M(t) L(t)
I(t)
Mature Parasites Juvenile Parasites
Immunity
Maturation
Gain of immunity Loss of immunity
Deathduetoimmunity
Naturaldeath
Naturaldeath
Invisible
Invisible
Invisible
γL(t)
νL(t) µI I(t)
βI(t)L(t)
µLL(t)
µMM(t)
Chris Drovandi QUT Seminar 2014
37. Bayesian Indirect Likelihood pBII Methods Examples References
Auxiliary Beta-Binomial model
The data show too much variation for Binomial
A Beta-Binomial model has an extra parameter to capture
dispersion
p(mi |αi , βi ) =
li
mi
B(mi + αi , li − mi + βi )
B(αi , βi )
,
Useful reparameterisation pi = αi /(αi + βi ) and
θi = 1/(αi + βi )
Relate the proportion and over dispersion parameters to time,
ti , and initial larvae, li , covariates
logit(pi ) = β0 + β1 log(ti ) + β2(log(ti ))2
,
log(θi ) =
η100, if li ≈ 100
η200, if li ≈ 200
,
Five parameters φ = (β0, β1, β2, η100, η200)
Chris Drovandi QUT Seminar 2014
40. Bayesian Indirect Likelihood pBII Methods Examples References
pBII results (ABC has regression adjustment)
0 1 2
x 10
−3
0
500
1000
1500
2000
ABC IS REG
ABC IP REG
ABC IL REG
pdBIL n=50
(y) ν
0 0.005 0.01 0.015 0.02 0.025
0
50
100
150
(z) µL
Chris Drovandi QUT Seminar 2014
41. Bayesian Indirect Likelihood pBII Methods Examples References
Discussion
pdBIL very different to ABC II theoretically
pdBIL needs to have a good auxiliary model. ABC II might be
useful is auxiliary model mis-specified
ABC II more flexible; can incorporate other summary statistics
Chris Drovandi QUT Seminar 2014
42. Bayesian Indirect Likelihood pBII Methods Examples References
Key References
Drovandi et al. (2011). Approximate Bayesian Computation
using Indirect Inference. JRSS C.
Drovandi, Pettitt and Lee (2014). Bayesian Indirect Inference
using a Parametric Auxiliary Model.
http://eprints.qut.edu.au/63767/
Gleim and Pigorsch (2013). Approximate Bayesian
Computation with Indirect Summary Statistics.
Gallant and McCullogh (2009). On the determination of
general scientific models with application to asset pricing.
JASA.
Reeves and Pettitt (2005). A theoretical framework for
approximate Bayesian computation. 20th IWSM.
Chris Drovandi QUT Seminar 2014