Pittsburgh and Toronto "Halloween US trip" seminars

  • 458 views
Uploaded on

Those are the slides of the talks given in Pittsburgh and Toronto, on Oct. 30 and 31, 2013

Those are the slides of the talks given in Pittsburgh and Toronto, on Oct. 30 and 31, 2013

More in: Education , Technology
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
458
On Slideshare
0
From Embeds
0
Number of Embeds
9

Actions

Shares
Downloads
22
Comments
0
Likes
0

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. Relevant statistics for (approximate) Bayesian model choice Christian P. Robert Universit´ Paris-Dauphine, Paris & University of Warwick, Coventry e bayesianstatistics@gmail.com
  • 2. a meeting of potential interest?! MCMSki IV to be held in Chamonix Mt Blanc, France, from Monday, Jan. 6 to Wed., Jan. 8 (and BNP workshop on Jan. 9), 2014 All aspects of MCMC++ theory and methodology and applications and more! All posters welcomed! Website http://www.pages.drexel.edu/∼mwl25/mcmski/
  • 3. Outline Intractable likelihoods ABC methods ABC as an inference machine ABC for model choice Model choice consistency Conclusion and perspectives
  • 4. Intractable likelihood Case of a well-defined statistical model where the likelihood function (θ|y) = f (y1 , . . . , yn |θ) is (really!) not available in closed form cannot (easily!) be either completed or demarginalised cannot be (at all!) estimated by an unbiased estimator c Prohibits direct implementation of a generic MCMC algorithm like Metropolis–Hastings
  • 5. Intractable likelihood Case of a well-defined statistical model where the likelihood function (θ|y) = f (y1 , . . . , yn |θ) is (really!) not available in closed form cannot (easily!) be either completed or demarginalised cannot be (at all!) estimated by an unbiased estimator c Prohibits direct implementation of a generic MCMC algorithm like Metropolis–Hastings
  • 6. Getting approximative Case of a well-defined statistical model where the likelihood function (θ|y) = f (y1 , . . . , yn |θ) is out of reach Empirical approximations to the original B problem Degrading the data precision down to a tolerance ε Replacing the likelihood with a non-parametric approximation Summarising/replacing the data with insufficient statistics
  • 7. Getting approximative Case of a well-defined statistical model where the likelihood function (θ|y) = f (y1 , . . . , yn |θ) is out of reach Empirical approximations to the original B problem Degrading the data precision down to a tolerance ε Replacing the likelihood with a non-parametric approximation Summarising/replacing the data with insufficient statistics
  • 8. Getting approximative Case of a well-defined statistical model where the likelihood function (θ|y) = f (y1 , . . . , yn |θ) is out of reach Empirical approximations to the original B problem Degrading the data precision down to a tolerance ε Replacing the likelihood with a non-parametric approximation Summarising/replacing the data with insufficient statistics
  • 9. Different [levels of] worries about abc Impact on B inference a mere computational issue (that will eventually end up being solved by more powerful computers, &tc, even if too costly in the short term, as for regular Monte Carlo methods) Not! an inferential issue (opening opportunities for new inference machine on complex/Big Data models, with legitimity differing from classical B approach) a Bayesian conundrum (how closely related to the/a B approach? more than recycling B tools? true B with raw data? a new type of empirical B inference?)
  • 10. Different [levels of] worries about abc Impact on B inference a mere computational issue (that will eventually end up being solved by more powerful computers, &tc, even if too costly in the short term, as for regular Monte Carlo methods) Not! an inferential issue (opening opportunities for new inference machine on complex/Big Data models, with legitimity differing from classical B approach) a Bayesian conundrum (how closely related to the/a B approach? more than recycling B tools? true B with raw data? a new type of empirical B inference?)
  • 11. Different [levels of] worries about abc Impact on B inference a mere computational issue (that will eventually end up being solved by more powerful computers, &tc, even if too costly in the short term, as for regular Monte Carlo methods) Not! an inferential issue (opening opportunities for new inference machine on complex/Big Data models, with legitimity differing from classical B approach) a Bayesian conundrum (how closely related to the/a B approach? more than recycling B tools? true B with raw data? a new type of empirical B inference?)
  • 12. Approximate Bayesian computation Intractable likelihoods ABC methods Genesis of ABC ABC basics Advances and interpretations ABC as knn ABC as an inference machine ABC for model choice Model choice consistency Conclusion and perspectives
  • 13. Genetic background of ABC skip genetics ABC is a recent computational technique that only requires being able to sample from the likelihood f (·|θ) This technique stemmed from population genetics models, about 15 years ago, and population geneticists still contribute significantly to methodological developments of ABC. [Griffith & al., 1997; Tavar´ & al., 1999] e
  • 14. Demo-genetic inference Each model is characterized by a set of parameters θ that cover historical (time divergence, admixture time ...), demographics (population sizes, admixture rates, migration rates, ...) and genetic (mutation rate, ...) factors The goal is to estimate these parameters from a dataset of polymorphism (DNA sample) y observed at the present time Problem: most of the time, we cannot calculate the likelihood of the polymorphism data f (y|θ)...
  • 15. Demo-genetic inference Each model is characterized by a set of parameters θ that cover historical (time divergence, admixture time ...), demographics (population sizes, admixture rates, migration rates, ...) and genetic (mutation rate, ...) factors The goal is to estimate these parameters from a dataset of polymorphism (DNA sample) y observed at the present time Problem: most of the time, we cannot calculate the likelihood of the polymorphism data f (y|θ)...
  • 16. Neutral model at a given microsatellite locus, in a closed panmictic population at equilibrium Sample of 8 genes Mutations according to the Simple stepwise Mutation Model (SMM) • date of the mutations ∼ Poisson process with intensity θ/2 over the branches • MRCA = 100 • independent mutations: ±1 with pr. 1/2
  • 17. Neutral model at a given microsatellite locus, in a closed panmictic population at equilibrium Kingman’s genealogy When time axis is normalized, T (k) ∼ Exp(k(k − 1)/2) Mutations according to the Simple stepwise Mutation Model (SMM) • date of the mutations ∼ Poisson process with intensity θ/2 over the branches • MRCA = 100 • independent mutations: ±1 with pr. 1/2
  • 18. Neutral model at a given microsatellite locus, in a closed panmictic population at equilibrium Kingman’s genealogy When time axis is normalized, T (k) ∼ Exp(k(k − 1)/2) Mutations according to the Simple stepwise Mutation Model (SMM) • date of the mutations ∼ Poisson process with intensity θ/2 over the branches • MRCA = 100 • independent mutations: ±1 with pr. 1/2
  • 19. Neutral model at a given microsatellite locus, in a closed panmictic population at equilibrium Kingman’s genealogy When time axis is normalized, T (k) ∼ Exp(k(k − 1)/2) Observations: leafs of the tree ^ θ=? Mutations according to the Simple stepwise Mutation Model (SMM) • date of the mutations ∼ Poisson process with intensity θ/2 over the branches • MRCA = 100 • independent mutations: ±1 with pr. 1/2
  • 20. Much more interesting models. . . several independent locus Independent gene genealogies and mutations different populations linked by an evolutionary scenario made of divergences, admixtures, migrations between populations, selection pressure, etc. larger sample size usually between 50 and 100 genes MRCA τ2 τ1 A typical evolutionary scenario: POP 0 POP 1 POP 2
  • 21. Instance of ecological questions How the Asian ladybird beetle arrived in Europe? Why does they swarm right now? What are the routes of invasion? How to get rid of them? Why did the chicken cross the road? [Lombaert & al., 2010, PLoS ONE]
  • 22. Worldwide invasion routes of Harmonia axyridis Worldwide routes of invasion of Harmonia axyridis For each outbreak, the arrow indicates the most likely invasion pathway and the associated posterior probability, with 95% credible intervals in brackets [Estoup et al., 2012, Molecular Ecology Res.]
  • 23. c Intractable likelihood Missing (too much missing!) data structure: f (y|θ) = G f (y|G , θ)f (G |θ)dG cannot be computed in a manageable way... [Stephens & Donnelly, 2000] The genealogies are considered as nuisance parameters This modelling clearly differs from the phylogenetic perspective where the tree is the parameter of interest.
  • 24. c Intractable likelihood Missing (too much missing!) data structure: f (y|θ) = G f (y|G , θ)f (G |θ)dG cannot be computed in a manageable way... [Stephens & Donnelly, 2000] The genealogies are considered as nuisance parameters This modelling clearly differs from the phylogenetic perspective where the tree is the parameter of interest.
  • 25. A?B?C? A stands for approximate [wrong likelihood / pic?] B stands for Bayesian C stands for computation [producing a parameter sample]
  • 26. A?B?C? A stands for approximate [wrong likelihood / pic?] B stands for Bayesian C stands for computation [producing a parameter sample]
  • 27. A?B?C? ESS=75.93 ESS=76.87 0.2 0.5 −0.2 −0.2 0.2 0.6 0.0 0.2 0.0 0.2 0.4 0.6 3 4 θ ESS=83.77 Density 0 0.0 0.2 0.4 0.6 −0.6 −0.4 θ ESS=92.42 −0.2 0.0 0.2 0.4 0.4 0.6 3.0 θ ESS=95.01 Density −0.4 0.0 0.2 0.4 0.6 −0.4 θ ESS=99.33 0.0 0.2 Density 3 θ ESS=87.28 0 Density 2.0 1.0 −0.6 −0.2 θ ESS=85.13 0.0 Density 2.0 1.0 0.0 θ ESS=139.2 3 0.6 2.0 Density −0.4 0.0 −0.5 2 Density 0 0.4 0.0 1.0 0.0 1.0 2.0 3.0 0.5 0.0 1.0 2.0 3.0 0.0 θ ESS=155.7 Density 0.2 1.0 2.0 1.0 Density 0.0 −0.5 Density 0.0 θ ESS=96.31 0.0 C stands for computation [producing a parameter sample] −0.4 Density −0.4 θ ESS=149.1 B stands for Bayesian 0.4 2 0.0 0.2 1 −0.2 0.0 θ ESS=108.4 1.5 Density −0.4 −0.2 0.0 1.0 2.0 3.0 −0.4 2 4 3 2 1 −0.6 1 2.0 1.0 0 Density 0.5 1 0.0 θ ESS=91.54 0.0 1.0 2.0 3.0 −0.5 A stands for approximate [wrong likelihood / pic?] 1.0 Density 0.0 1.0 0.0 Density 4 ESS=155.6 −0.4 −0.2 0.0 0.2 0.4 −0.2 0.0 0.2 0.4 0.6
  • 28. How Bayesian is aBc? Could we turn the resolution into a Bayesian answer? ideally so (not meaningfull: requires ∞-ly powerful computer approximation error unknown (w/o costly simulation) true B for wrong model (formal and artificial) true B for noisy model (much more convincing) true B for estimated likelihood (back to econometrics?) illuminating the tension between information and precision
  • 29. ABC methodology Bayesian setting: target is π(θ)f (x|θ) When likelihood f (x|θ) not in closed form, likelihood-free rejection technique: Foundation For an observation y ∼ f (y|θ), under the prior π(θ), if one keeps jointly simulating θ ∼ π(θ) , z ∼ f (z|θ ) , until the auxiliary variable z is equal to the observed value, z = y, then the selected θ ∼ π(θ|y) [Rubin, 1984; Diggle & Gratton, 1984; Tavar´ et al., 1997] e
  • 30. ABC methodology Bayesian setting: target is π(θ)f (x|θ) When likelihood f (x|θ) not in closed form, likelihood-free rejection technique: Foundation For an observation y ∼ f (y|θ), under the prior π(θ), if one keeps jointly simulating θ ∼ π(θ) , z ∼ f (z|θ ) , until the auxiliary variable z is equal to the observed value, z = y, then the selected θ ∼ π(θ|y) [Rubin, 1984; Diggle & Gratton, 1984; Tavar´ et al., 1997] e
  • 31. ABC methodology Bayesian setting: target is π(θ)f (x|θ) When likelihood f (x|θ) not in closed form, likelihood-free rejection technique: Foundation For an observation y ∼ f (y|θ), under the prior π(θ), if one keeps jointly simulating θ ∼ π(θ) , z ∼ f (z|θ ) , until the auxiliary variable z is equal to the observed value, z = y, then the selected θ ∼ π(θ|y) [Rubin, 1984; Diggle & Gratton, 1984; Tavar´ et al., 1997] e
  • 32. A as A...pproximative When y is a continuous random variable, strict equality z = y is replaced with a tolerance zone ρ(y, z) where ρ is a distance Output distributed from def π(θ) Pθ {ρ(y, z) < } ∝ π(θ|ρ(y, z) < ) [Pritchard et al., 1999]
  • 33. A as A...pproximative When y is a continuous random variable, strict equality z = y is replaced with a tolerance zone ρ(y, z) where ρ is a distance Output distributed from def π(θ) Pθ {ρ(y, z) < } ∝ π(θ|ρ(y, z) < ) [Pritchard et al., 1999]
  • 34. ABC algorithm In most implementations, further degree of A...pproximation: Algorithm 1 Likelihood-free rejection sampler for i = 1 to N do repeat generate θ from the prior distribution π(·) generate z from the likelihood f (·|θ ) until ρ{η(z), η(y)} set θi = θ end for where η(y) defines a (not necessarily sufficient) statistic
  • 35. Output The likelihood-free algorithm samples from the marginal in z of: π (θ, z|y) = where A ,y π(θ)f (z|θ)IA ,y (z) , A ,y ×Θ π(θ)f (z|θ)dzdθ = {z ∈ D|ρ(η(z), η(y)) < }. The idea behind ABC is that the summary statistics coupled with a small tolerance should provide a good approximation of the posterior distribution: π (θ|y) = π (θ, z|y)dz ≈ π(θ|y) . ...does it?!
  • 36. Output The likelihood-free algorithm samples from the marginal in z of: π (θ, z|y) = where A ,y π(θ)f (z|θ)IA ,y (z) , A ,y ×Θ π(θ)f (z|θ)dzdθ = {z ∈ D|ρ(η(z), η(y)) < }. The idea behind ABC is that the summary statistics coupled with a small tolerance should provide a good approximation of the posterior distribution: π (θ|y) = π (θ, z|y)dz ≈ π(θ|y) . ...does it?!
  • 37. Output The likelihood-free algorithm samples from the marginal in z of: π (θ, z|y) = where A ,y π(θ)f (z|θ)IA ,y (z) , A ,y ×Θ π(θ)f (z|θ)dzdθ = {z ∈ D|ρ(η(z), η(y)) < }. The idea behind ABC is that the summary statistics coupled with a small tolerance should provide a good approximation of the posterior distribution: π (θ|y) = π (θ, z|y)dz ≈ π(θ|y) . ...does it?!
  • 38. Output The likelihood-free algorithm samples from the marginal in z of: π (θ, z|y) = where A ,y π(θ)f (z|θ)IA ,y (z) , A ,y ×Θ π(θ)f (z|θ)dzdθ = {z ∈ D|ρ(η(z), η(y)) < }. The idea behind ABC is that the summary statistics coupled with a small tolerance should provide a good approximation of the restricted posterior distribution: π (θ|y) = π (θ, z|y)dz ≈ π(θ|η(y)) . Not so good..!
  • 39. Comments Role of distance paramount (because = 0) Scaling of components of η(y) is also determinant matters little if “small enough” representative of “curse of dimensionality” small is beautiful! the data as a whole may be paradoxically weakly informative for ABC
  • 40. curse of dimensionality in summaries Consider that the simulated summary statistics η(y1 ), . . . , η(yN ) the observed summary statistics η(yobs ) are iid with uniform distribution on [0, 1]d Take δ∞ (d, N) = E δ∞ (1, N) δ∞ (2, N) δ∞ (10, N) δ∞ (200, N) min i=1,...,N N = 100 0.0025 0.033 0.28 0.48 η(yobs ) − η(yi ) N = 1, 000 0.00025 0.01 0.22 0.48 ∞ N = 10, 000 0.000025 0.0033 0.18 0.47 N = 100, 000 0.0000025 0.001 0.14 0.46
  • 41. ABC (simul’) advances how approximative is ABC? ABC as knn Simulating from the prior is often poor in efficiency Either modify the proposal distribution on θ to increase the density of x’s within the vicinity of y ... [Marjoram et al, 2003; Bortot et al., 2007, Sisson et al., 2007] ...or by viewing the problem as a conditional density estimation and by developing techniques to allow for larger [Beaumont et al., 2002] .....or even by including in the inferential framework [ABCµ ] [Ratmann et al., 2009]
  • 42. ABC (simul’) advances how approximative is ABC? ABC as knn Simulating from the prior is often poor in efficiency Either modify the proposal distribution on θ to increase the density of x’s within the vicinity of y ... [Marjoram et al, 2003; Bortot et al., 2007, Sisson et al., 2007] ...or by viewing the problem as a conditional density estimation and by developing techniques to allow for larger [Beaumont et al., 2002] .....or even by including in the inferential framework [ABCµ ] [Ratmann et al., 2009]
  • 43. ABC (simul’) advances how approximative is ABC? ABC as knn Simulating from the prior is often poor in efficiency Either modify the proposal distribution on θ to increase the density of x’s within the vicinity of y ... [Marjoram et al, 2003; Bortot et al., 2007, Sisson et al., 2007] ...or by viewing the problem as a conditional density estimation and by developing techniques to allow for larger [Beaumont et al., 2002] .....or even by including in the inferential framework [ABCµ ] [Ratmann et al., 2009]
  • 44. ABC (simul’) advances how approximative is ABC? ABC as knn Simulating from the prior is often poor in efficiency Either modify the proposal distribution on θ to increase the density of x’s within the vicinity of y ... [Marjoram et al, 2003; Bortot et al., 2007, Sisson et al., 2007] ...or by viewing the problem as a conditional density estimation and by developing techniques to allow for larger [Beaumont et al., 2002] .....or even by including in the inferential framework [ABCµ ] [Ratmann et al., 2009]
  • 45. ABC as knn [Biau et al., 2013, Annales de l’IHP] Practice of ABC: determine tolerance distances, say 10% or 1% quantile, = N as a quantile on observed = qα (d1 , . . . , dN ) Interpretation of ε as nonparametric bandwidth only approximation of the actual practice [Blum & Fran¸ois, 2010] c ABC is a k-nearest neighbour (knn) method with kN = N N [Loftsgaarden & Quesenberry, 1965]
  • 46. ABC as knn [Biau et al., 2013, Annales de l’IHP] Practice of ABC: determine tolerance distances, say 10% or 1% quantile, = N as a quantile on observed = qα (d1 , . . . , dN ) Interpretation of ε as nonparametric bandwidth only approximation of the actual practice [Blum & Fran¸ois, 2010] c ABC is a k-nearest neighbour (knn) method with kN = N N [Loftsgaarden & Quesenberry, 1965]
  • 47. ABC as knn [Biau et al., 2013, Annales de l’IHP] Practice of ABC: determine tolerance distances, say 10% or 1% quantile, = N as a quantile on observed = qα (d1 , . . . , dN ) Interpretation of ε as nonparametric bandwidth only approximation of the actual practice [Blum & Fran¸ois, 2010] c ABC is a k-nearest neighbour (knn) method with kN = N N [Loftsgaarden & Quesenberry, 1965]
  • 48. ABC consistency Provided kN / log log N −→ ∞ and kN /N −→ 0 as N → ∞, for almost all s0 (with respect to the distribution of S), with probability 1, 1 kN kN ϕ(θj ) −→ E[ϕ(θj )|S = s0 ] j=1 [Devroye, 1982] Biau et al. (2013) also recall pointwise and integrated mean square error consistency results on the corresponding kernel estimate of the conditional posterior distribution, under constraints kN → ∞, kN /N → 0, hN → 0 p and hN kN → ∞,
  • 49. ABC consistency Provided kN / log log N −→ ∞ and kN /N −→ 0 as N → ∞, for almost all s0 (with respect to the distribution of S), with probability 1, 1 kN kN ϕ(θj ) −→ E[ϕ(θj )|S = s0 ] j=1 [Devroye, 1982] Biau et al. (2013) also recall pointwise and integrated mean square error consistency results on the corresponding kernel estimate of the conditional posterior distribution, under constraints kN → ∞, kN /N → 0, hN → 0 p and hN kN → ∞,
  • 50. Rates of convergence Further assumptions (on target and kernel) allow for precise (integrated mean square) convergence rates (as a power of the sample size N), derived from classical k-nearest neighbour regression, like 4 when m = 1, 2, 3, kN ≈ N (p+4)/(p+8) and rate N − p+8 4 when m = 4, kN ≈ N (p+4)/(p+8) and rate N − p+8 log N 4 when m > 4, kN ≈ N (p+4)/(m+p+4) and rate N − m+p+4 [Biau et al., 2013] Drag: Only applies to sufficient summary statistics
  • 51. Rates of convergence Further assumptions (on target and kernel) allow for precise (integrated mean square) convergence rates (as a power of the sample size N), derived from classical k-nearest neighbour regression, like 4 when m = 1, 2, 3, kN ≈ N (p+4)/(p+8) and rate N − p+8 4 when m = 4, kN ≈ N (p+4)/(p+8) and rate N − p+8 log N 4 when m > 4, kN ≈ N (p+4)/(m+p+4) and rate N − m+p+4 [Biau et al., 2013] Drag: Only applies to sufficient summary statistics
  • 52. ABC inference machine Intractable likelihoods ABC methods ABC as an inference machine Error inc. Exact BC and approximate targets summary statistic ABC for model choice Model choice consistency Conclusion and perspectives
  • 53. How Bayesian is aBc..? may be a convergent method of inference (meaningful? sufficient? foreign?) approximation error unknown (w/o massive simulation) pragmatic/empirical B (there is no other solution!) many calibration issues (tolerance, distance, statistics) the NP side should be incorporated into the whole B picture the approximation error should also be part of the B inference
  • 54. ABCµ Idea Infer about the error as well as about the parameter: Use of a joint density f (θ, |y) ∝ ξ( |y, θ) × πθ (θ) × π ( ) where y is the data, and ξ( |y, θ) is the prior predictive density of ρ(η(z), η(y)) given θ and y when z ∼ f (z|θ) Warning! Replacement of ξ( |y, θ) with a non-parametric kernel approximation. [Ratmann, Andrieu, Wiuf and Richardson, 2009, PNAS]
  • 55. ABCµ Idea Infer about the error as well as about the parameter: Use of a joint density f (θ, |y) ∝ ξ( |y, θ) × πθ (θ) × π ( ) where y is the data, and ξ( |y, θ) is the prior predictive density of ρ(η(z), η(y)) given θ and y when z ∼ f (z|θ) Warning! Replacement of ξ( |y, θ) with a non-parametric kernel approximation. [Ratmann, Andrieu, Wiuf and Richardson, 2009, PNAS]
  • 56. ABCµ Idea Infer about the error as well as about the parameter: Use of a joint density f (θ, |y) ∝ ξ( |y, θ) × πθ (θ) × π ( ) where y is the data, and ξ( |y, θ) is the prior predictive density of ρ(η(z), η(y)) given θ and y when z ∼ f (z|θ) Warning! Replacement of ξ( |y, θ) with a non-parametric kernel approximation. [Ratmann, Andrieu, Wiuf and Richardson, 2009, PNAS]
  • 57. Wilkinson’s exact BC (not exactly!) ABC approximation error (i.e. non-zero tolerance) replaced with exact simulation from a controlled approximation to the target, convolution of true posterior with kernel function π (θ, z|y) = π(θ)f (z|θ)K (y − z) , π(θ)f (z|θ)K (y − z)dzdθ with K kernel parameterised by bandwidth . [Wilkinson, 2013] Theorem The ABC algorithm based on the production of a randomised ˜ observation y = y + ξ, ξ ∼ K , and an acceptance probability of K (y − z)/M gives draws from the posterior distribution π(θ|y).
  • 58. Wilkinson’s exact BC (not exactly!) ABC approximation error (i.e. non-zero tolerance) replaced with exact simulation from a controlled approximation to the target, convolution of true posterior with kernel function π (θ, z|y) = π(θ)f (z|θ)K (y − z) , π(θ)f (z|θ)K (y − z)dzdθ with K kernel parameterised by bandwidth . [Wilkinson, 2013] Theorem The ABC algorithm based on the production of a randomised ˜ observation y = y + ξ, ξ ∼ K , and an acceptance probability of K (y − z)/M gives draws from the posterior distribution π(θ|y).
  • 59. How exact a BC? Pros Pseudo-data from true model vs. observed data from noisy model Outcome is completely controlled Link with ABCµ when assuming y observed with measurement error with density K Relates to the theory of model approximation [Kennedy & O’Hagan, 2001] leads to “noisy ABC”: just do it!, i.e. perturbates the data down to precision ε and proceed [Fearnhead & Prangle, 2012] Cons Requires K to be bounded by M True approximation error never assessed
  • 60. How exact a BC? Pros Pseudo-data from true model vs. observed data from noisy model Outcome is completely controlled Link with ABCµ when assuming y observed with measurement error with density K Relates to the theory of model approximation [Kennedy & O’Hagan, 2001] leads to “noisy ABC”: just do it!, i.e. perturbates the data down to precision ε and proceed [Fearnhead & Prangle, 2012] Cons Requires K to be bounded by M True approximation error never assessed
  • 61. Which summary? Fundamental difficulty of the choice of the summary statistic when there is no non-trivial sufficient statistics Loss of statistical information balanced against gain in data roughening Approximation error and information loss remain unknown Choice of statistics induces choice of distance function towards standardisation may be imposed for external/practical reasons (e.g., DIYABC) may gather several non-B point estimates can learn about efficient combination
  • 62. Which summary? Fundamental difficulty of the choice of the summary statistic when there is no non-trivial sufficient statistics Loss of statistical information balanced against gain in data roughening Approximation error and information loss remain unknown Choice of statistics induces choice of distance function towards standardisation may be imposed for external/practical reasons (e.g., DIYABC) may gather several non-B point estimates can learn about efficient combination
  • 63. Which summary? Fundamental difficulty of the choice of the summary statistic when there is no non-trivial sufficient statistics Loss of statistical information balanced against gain in data roughening Approximation error and information loss remain unknown Choice of statistics induces choice of distance function towards standardisation may be imposed for external/practical reasons (e.g., DIYABC) may gather several non-B point estimates can learn about efficient combination
  • 64. Semi-automatic ABC Fearnhead and Prangle (2012) study ABC and the selection of the summary statistic in close proximity to Wilkinson’s proposal ABC considered as inferential method and calibrated as such randomised (or ‘noisy’) version of the summary statistics ˜ η(y) = η(y) + τ derivation of a well-calibrated version of ABC, i.e. an algorithm that gives proper predictions for the distribution associated with this randomised summary statistic [weak property]
  • 65. Summary [of F&P/statistics) optimality of the posterior expectation E[θ|y] of the parameter of interest as summary statistics η(y)! [requires iterative process] use of the standard quadratic loss function (θ − θ0 )T A(θ − θ0 ) recent extension to model choice, optimality of Bayes factor B12 (y) [F&P, ISBA 2012, Kyoto]
  • 66. Summary [of F&P/statistics) optimality of the posterior expectation E[θ|y] of the parameter of interest as summary statistics η(y)! [requires iterative process] use of the standard quadratic loss function (θ − θ0 )T A(θ − θ0 ) recent extension to model choice, optimality of Bayes factor B12 (y) [F&P, ISBA 2012, Kyoto]
  • 67. LDA summaries for model choice In parallel to F& P semi-automatic ABC, selection of most discriminant subvector out of a collection of summary statistics, based on Linear Discriminant Analysis (LDA) [Estoup & al., 2012, Mol. Ecol. Res.] Solution now implemented in DIYABC.2 [Cornuet & al., 2008, Bioinf.; Estoup & al., 2013]
  • 68. Implementation Step 1: Rake a subset of the α% (e.g., 1%) best simulations from an ABC reference table usually including 106 –109 simulations for each of the M compared scenarios/models Selection based on normalized Euclidian distance computed between observed and simulated raw summaries Step 2: run LDA on this subset to transform summaries into (M − 1) discriminant variables When computing LDA functions, weight simulated data with an Epanechnikov kernel Step 3: Estimation of the posterior probabilities of each competing scenario/model by polychotomous local logistic regression against the M − 1 most discriminant variables [Cornuet & al., 2008, Bioinformatics]
  • 69. LDA advantages much faster computation of scenario probabilities via polychotomous regression a (much) lower number of explanatory variables improves the accuracy of the ABC approximation, reduces the tolerance and avoids extra costs in constructing the reference table allows for a large collection of initial summaries ability to evaluate Type I and Type II errors on more complex models LDA reduces correlation among explanatory variables When available, using both simulated and real data sets, posterior probabilities of scenarios computed from LDA-transformed and raw summaries are strongly correlated
  • 70. LDA advantages much faster computation of scenario probabilities via polychotomous regression a (much) lower number of explanatory variables improves the accuracy of the ABC approximation, reduces the tolerance and avoids extra costs in constructing the reference table allows for a large collection of initial summaries ability to evaluate Type I and Type II errors on more complex models LDA reduces correlation among explanatory variables When available, using both simulated and real data sets, posterior probabilities of scenarios computed from LDA-transformed and raw summaries are strongly correlated
  • 71. ABC for model choice Intractable likelihoods ABC methods ABC as an inference machine ABC for model choice BMC Principle Gibbs random fields (counterexample) Generic ABC model choice Model choice consistency Conclusion and perspectives
  • 72. Bayesian model choice BMC Principle Several models M1 , M2 , . . . are considered simultaneously for dataset y and model index M central to inference. Use of prior π(M = m), plus prior distribution on the parameter conditional on the value m of the model index, πm (θm )
  • 73. Bayesian model choice BMC Principle Several models M1 , M2 , . . . are considered simultaneously for dataset y and model index M central to inference. Goal is to derive the posterior distribution of M, π(M = m|data) a challenging computational target when models are complex.
  • 74. Generic ABC for model choice Algorithm 2 Likelihood-free model choice sampler (ABC-MC) for t = 1 to T do repeat Generate m from the prior π(M = m) Generate θm from the prior πm (θm ) Generate z from the model fm (z|θm ) until ρ{η(z), η(y)} < Set m(t) = m and θ(t) = θm end for [Grelaud & al., 2009; Toni & al., 2009]
  • 75. ABC estimates Posterior probability π(M = m|y) approximated by the frequency of acceptances from model m 1 T T Im(t) =m . t=1
  • 76. ABC estimates Posterior probability π(M = m|y) approximated by the frequency of acceptances from model m 1 T T Im(t) =m . t=1 Extension to a weighted polychotomous logistic regression estimate of π(M = m|y), with non-parametric kernel weights [Cornuet et al., DIYABC, 2009]
  • 77. Potts model Skip MRFs Potts model Distribution with an energy function of the form θS(y) = θ δyl =yi l∼i where l∼i denotes a neighbourhood structure In most realistic settings, summation Zθ = exp{θS(x)} x∈X involves too many terms to be manageable and numerical approximations cannot always be trusted
  • 78. Potts model Skip MRFs Potts model Distribution with an energy function of the form θS(y) = θ δyl =yi l∼i where l∼i denotes a neighbourhood structure In most realistic settings, summation Zθ = exp{θS(x)} x∈X involves too many terms to be manageable and numerical approximations cannot always be trusted
  • 79. Neighbourhood relations Setup Choice to be made between M neighbourhood relations m i ∼i m (0 M − 1) with Sm (x) = I{xi =xi } m i ∼i driven by the posterior probabilities of the models.
  • 80. Model index Computational target: P(M = m|x) ∝ fm (x|θm )πm (θm ) dθm π(M = m) Θm If S(x) sufficient statistic for the joint parameters (M, θ0 , . . . , θM−1 ), P(M = m|x) = P(M = m|S(x)) .
  • 81. Model index Computational target: P(M = m|x) ∝ fm (x|θm )πm (θm ) dθm π(M = m) Θm If S(x) sufficient statistic for the joint parameters (M, θ0 , . . . , θM−1 ), P(M = m|x) = P(M = m|S(x)) .
  • 82. Sufficient statistics in Gibbs random fields Each model m has its own sufficient statistic Sm (·) and S(·) = (S0 (·), . . . , SM−1 (·)) is also (model-)sufficient. Explanation: For Gibbs random fields, 1 2 x|M = m ∼ fm (x|θm ) = fm (x|S(x))fm (S(x)|θm ) 1 f 2 (S(x)|θm ) = n(S(x)) m where ˜ ˜ n(S(x)) = {x ∈ X : S(x) = S(x)} c S(x) is sufficient for the joint parameters
  • 83. Sufficient statistics in Gibbs random fields Each model m has its own sufficient statistic Sm (·) and S(·) = (S0 (·), . . . , SM−1 (·)) is also (model-)sufficient. Explanation: For Gibbs random fields, 1 2 x|M = m ∼ fm (x|θm ) = fm (x|S(x))fm (S(x)|θm ) 1 = f 2 (S(x)|θm ) n(S(x)) m where ˜ ˜ n(S(x)) = {x ∈ X : S(x) = S(x)} c S(x) is sufficient for the joint parameters
  • 84. Toy example iid Bernoulli model versus two-state first-order Markov chain, i.e. n f0 (x|θ0 ) = exp θ0 I{xi =1} {1 + exp(θ0 )}n , i=1 versus f1 (x|θ1 ) = 1 exp θ1 2 n I{xi =xi−1 } {1 + exp(θ1 )}n−1 , i=2 with priors θ0 ∼ U(−5, 5) and θ1 ∼ U(0, 6) (inspired by “phase transition” boundaries).
  • 85. About sufficiency If η1 (x) sufficient statistic for model m = 1 and parameter θ1 and η2 (x) sufficient statistic for model m = 2 and parameter θ2 , (η1 (x), η2 (x)) is not always sufficient for (m, θm ) c Potential loss of information at the testing level
  • 86. About sufficiency If η1 (x) sufficient statistic for model m = 1 and parameter θ1 and η2 (x) sufficient statistic for model m = 2 and parameter θ2 , (η1 (x), η2 (x)) is not always sufficient for (m, θm ) c Potential loss of information at the testing level
  • 87. Poisson/geometric example Sample x = (x1 , . . . , xn ) from either a Poisson P(λ) or from a geometric G(p) Sum n S= xi = η(x) i=1 sufficient statistic for either model but not simultaneously
  • 88. Limiting behaviour of B12 (T → ∞) ABC approximation B12 (y) = T t=1 Imt =1 Iρ{η(zt ),η(y)} T t=1 Imt =2 Iρ{η(zt ),η(y)} , where the (mt , z t )’s are simulated from the (joint) prior As T goes to infinity, limit B12 (y) = = Iρ{η(z),η(y)} Iρ{η(z),η(y)} Iρ{η,η(y)} Iρ{η,η(y)} π1 (θ1 )f1 (z|θ1 ) dz dθ1 π2 (θ2 )f2 (z|θ2 ) dz dθ2 π1 (θ1 )f1η (η|θ1 ) dη dθ1 , π2 (θ2 )f2η (η|θ2 ) dη dθ2 where f1η (η|θ1 ) and f2η (η|θ2 ) distributions of η(z)
  • 89. Limiting behaviour of B12 (T → ∞) ABC approximation B12 (y) = T t=1 Imt =1 Iρ{η(zt ),η(y)} T t=1 Imt =2 Iρ{η(zt ),η(y)} , where the (mt , z t )’s are simulated from the (joint) prior As T goes to infinity, limit B12 (y) = = Iρ{η(z),η(y)} Iρ{η(z),η(y)} Iρ{η,η(y)} Iρ{η,η(y)} π1 (θ1 )f1 (z|θ1 ) dz dθ1 π2 (θ2 )f2 (z|θ2 ) dz dθ2 π1 (θ1 )f1η (η|θ1 ) dη dθ1 , π2 (θ2 )f2η (η|θ2 ) dη dθ2 where f1η (η|θ1 ) and f2η (η|θ2 ) distributions of η(z)
  • 90. Limiting behaviour of B12 ( → 0) When goes to zero, η B12 (y) = π1 (θ1 )f1η (η(y)|θ1 ) dθ1 π2 (θ2 )f2η (η(y)|θ2 ) dθ2 c Bayes factor based on the sole observation of η(y)
  • 91. Limiting behaviour of B12 ( → 0) When goes to zero, η B12 (y) = π1 (θ1 )f1η (η(y)|θ1 ) dθ1 π2 (θ2 )f2η (η(y)|θ2 ) dθ2 c Bayes factor based on the sole observation of η(y)
  • 92. Limiting behaviour of B12 (under sufficiency) If η(y) sufficient statistic in both models, fi (y|θi ) = gi (y)fi η (η(y)|θi ) Thus = Θ1 π(θ1 )g1 (y)f1η (η(y)|θ1 ) dθ1 Θ2 B12 (y) = π(θ2 )g2 (y)f2η (η(y)|θ2 ) dθ2 g1 (y) g2 (y) π1 (θ1 )f1η (η(y)|θ1 ) dθ1 g1 (y) η = B (y) . g2 (y) 12 π2 (θ2 )f2η (η(y)|θ2 ) dθ2 [Didelot, Everitt, Johansen & Lawson, 2011] c No discrepancy only when cross-model sufficiency
  • 93. Limiting behaviour of B12 (under sufficiency) If η(y) sufficient statistic in both models, fi (y|θi ) = gi (y)fi η (η(y)|θi ) Thus = Θ1 π(θ1 )g1 (y)f1η (η(y)|θ1 ) dθ1 Θ2 B12 (y) = π(θ2 )g2 (y)f2η (η(y)|θ2 ) dθ2 g1 (y) g2 (y) π1 (θ1 )f1η (η(y)|θ1 ) dθ1 g1 (y) η = B (y) . g2 (y) 12 π2 (θ2 )f2η (η(y)|θ2 ) dθ2 [Didelot, Everitt, Johansen & Lawson, 2011] c No discrepancy only when cross-model sufficiency
  • 94. Poisson/geometric example (back) Sample x = (x1 , . . . , xn ) from either a Poisson P(λ) or from a geometric G(p) Discrepancy ratio S!n−S / i xi ! g1 (x) = g2 (x) 1 n+S−1 S
  • 95. Poisson/geometric discrepancy η Range of B12 (x) versus B12 (x): The values produced have nothing in common.
  • 96. Formal recovery Creating an encompassing exponential family f (x|θ1 , θ2 , α1 , α2 ) ∝ exp{θT η1 (x) + θT η2 (x) + α1 t1 (x) + α2 t2 (x)} 1 2 leads to a sufficient statistic (η1 (x), η2 (x), t1 (x), t2 (x)) [Didelot, Everitt, Johansen & Lawson, 2011]
  • 97. Formal recovery Creating an encompassing exponential family f (x|θ1 , θ2 , α1 , α2 ) ∝ exp{θT η1 (x) + θT η2 (x) + α1 t1 (x) + α2 t2 (x)} 1 2 leads to a sufficient statistic (η1 (x), η2 (x), t1 (x), t2 (x)) [Didelot, Everitt, Johansen & Lawson, 2011] In the Poisson/geometric case, if discrepancy i xi ! is added to S, no
  • 98. Formal recovery Creating an encompassing exponential family f (x|θ1 , θ2 , α1 , α2 ) ∝ exp{θT η1 (x) + θT η2 (x) + α1 t1 (x) + α2 t2 (x)} 1 2 leads to a sufficient statistic (η1 (x), η2 (x), t1 (x), t2 (x)) [Didelot, Everitt, Johansen & Lawson, 2011] Only applies in genuine sufficiency settings... c Inability to evaluate information loss due to summary statistics
  • 99. Meaning of the ABC-Bayes factor The ABC approximation to the Bayes Factor is based solely on the summary statistics.... In the Poisson/geometric case, if E[yi ] = θ0 > 0, η lim B12 (y) = n→∞ (θ0 + 1)2 −θ0 e θ0
  • 100. Meaning of the ABC-Bayes factor The ABC approximation to the Bayes Factor is based solely on the summary statistics.... In the Poisson/geometric case, if E[yi ] = θ0 > 0, η lim B12 (y) = n→∞ (θ0 + 1)2 −θ0 e θ0
  • 101. ABC model choice consistency Intractable likelihoods ABC methods ABC as an inference machine ABC for model choice Model choice consistency Formalised framework Consistency results Summary statistics Conclusion and perspectives
  • 102. The starting point Central question to the validation of ABC for model choice: When is a Bayes factor based on an insufficient statistic T(y) consistent? T Note/warnin: c drawn on T(y) through B12 (y) necessarily differs from c drawn on y through B12 (y) [Marin, Pillai, X, & Rousseau, JRSS B, 2013]
  • 103. The starting point Central question to the validation of ABC for model choice: When is a Bayes factor based on an insufficient statistic T(y) consistent? T Note/warnin: c drawn on T(y) through B12 (y) necessarily differs from c drawn on y through B12 (y) [Marin, Pillai, X, & Rousseau, JRSS B, 2013]
  • 104. A benchmark if toy example Comparison suggested by referee of PNAS paper [thanks!]: [X, Cornuet, Marin, & Pillai, Aug. 2011] Model M1 : y ∼ N(θ1 , 1) opposed √ to model M2 : y ∼ L(θ2 , 1/ 2), Laplace distribution with mean θ2 √ and scale parameter 1/ 2 (variance one).
  • 105. A benchmark if toy example Comparison suggested by referee of PNAS paper [thanks!]: [X, Cornuet, Marin, & Pillai, Aug. 2011] Model M1 : y ∼ N(θ1 , 1) opposed √ to model M2 : y ∼ L(θ2 , 1/ 2), Laplace distribution with mean θ2 √ and scale parameter 1/ 2 (variance one). Four possible statistics 1. sample mean y (sufficient for M1 if not M2 ); 2. sample median med(y) (insufficient); 3. sample variance var(y) (ancillary); 4. median absolute deviation mad(y) = med(|y − med(y)|);
  • 106. A benchmark if toy example Comparison suggested by referee of PNAS paper [thanks!]: [X, Cornuet, Marin, & Pillai, Aug. 2011] Model M1 : y ∼ N(θ1 , 1) opposed √ to model M2 : y ∼ L(θ2 , 1/ 2), Laplace distribution with mean θ2 √ and scale parameter 1/ 2 (variance one). 0.3 0.4 0.5 0.6 0.7 n=100 q q q 0.1 0.2 q q q q q 0.0 q q q Gauss Laplace
  • 107. A benchmark if toy example Comparison suggested by referee of PNAS paper [thanks!]: [X, Cornuet, Marin, & Pillai, Aug. 2011] Model M1 : y ∼ N(θ1 , 1) opposed √ to model M2 : y ∼ L(θ2 , 1/ 2), Laplace distribution with mean θ2 √ and scale parameter 1/ 2 (variance one). n=100 0.7 1.0 n=100 q 0.6 q 0.8 q q 0.5 q q q q q q q 0.4 q q q q 0.1 0.2 q q 0.2 0.3 0.4 0.6 q q q q q q 0.0 q q q Gauss Laplace 0.0 q Gauss Laplace
  • 108. Framework Starting from sample y = (y1 , . . . , yn ) the observed sample, not necessarily iid with true distribution y ∼ Pn Summary statistics T(y) = Tn = (T1 (y), T2 (y), · · · , Td (y)) ∈ Rd with true distribution Tn ∼ Gn .
  • 109. Framework c Comparison of – under M1 , y ∼ F1,n (·|θ1 ) where θ1 ∈ Θ1 ⊂ Rp1 – under M2 , y ∼ F2,n (·|θ2 ) where θ2 ∈ Θ2 ⊂ Rp2 turned into – under M1 , T(y) ∼ G1,n (·|θ1 ), and θ1 |T(y) ∼ π1 (·|Tn ) – under M2 , T(y) ∼ G2,n (·|θ2 ), and θ2 |T(y) ∼ π2 (·|Tn )
  • 110. Assumptions A collection of asymptotic “standard” assumptions: [A1] is a standard central limit theorem under the true model [A2] controls the large deviations of the estimator Tn from the estimand µ(θ) [A3] is the standard prior mass condition found in Bayesian asymptotics (di effective dimension of the parameter) [A4] restricts the behaviour of the model density against the true density [Think CLT!]
  • 111. Assumptions A collection of asymptotic “standard” assumptions: [Think CLT!] [A1] There exist a sequence {vn } converging to +∞, a distribution Q, a symmetric, d × d positive definite matrix V0 and a vector µ0 ∈ Rd such that −1/2 vn V0 (Tn − µ0 ) n→∞ Q, under Gn
  • 112. Assumptions A collection of asymptotic “standard” assumptions: [Think CLT!] [A2] For i = 1, 2, there exist sets Fn,i ⊂ Θi , functions µi (θi ) and constants i , τi , αi > 0 such that for all τ > 0, sup θi ∈Fn,i Gi,n |Tn − µi (θi )| > τ|µi (θi ) − µ0 | ∧ (|τµi (θi ) − µ0 | ∧ −αi i) with c −τ πi (Fn,i ) = o(vn i ). i |θi −α vn i
  • 113. Assumptions A collection of asymptotic “standard” assumptions: [Think CLT!] [A3] If inf{|µi (θi ) − µ0 |; θi ∈ Θi } = 0, defining (u > 0) Sn,i (u) = θi ∈ Fn,i ; |µi (θi ) − µ0 | −1 u vn , there exist constants di < τi ∧ αi − 1 such that −d πi (Sn,i (u)) ∼ u di vn i , ∀u vn
  • 114. Assumptions A collection of asymptotic “standard” assumptions: [Think CLT!] [A4] If inf{|µi (θi ) − µ0 |; θi ∈ Θi } = 0, for any U, δ > 0 and (En )n such that, if θi ∈ Sn,i (U) En ⊂ {t; gi (t|θi ) < δgn (t)} > 0, there exist c and Gn (En ) < .
  • 115. Assumptions A collection of asymptotic “standard” assumptions: [Think CLT!] Again (sumup) [A1] is a standard central limit theorem under the true model [A2] controls the large deviations of the estimator Tn from the estimand µ(θ) [A3] is the standard prior mass condition found in Bayesian asymptotics (di effective dimension of the parameter) [A4] restricts the behaviour of the model density against the true density
  • 116. Effective dimension Understanding di in [A3]: defined only when µ0 ∈ {µi (θi ), θi ∈ Θi }, πi (θi : |µi (θi ) − µ0 | < n−1/2 ) = O(n−di /2 ) is the effective dimension of the model Θi around µ0
  • 117. Effective dimension Understanding di in [A3]: when inf{|µi (θi ) − µ0 |; θi ∈ Θi } = 0, mi (Tn ) −d ∼ vn i , gn (Tn ) thus log(mi (Tn )/gn (Tn )) ∼ −di log vn −d and vn i penalization factor resulting from integrating θi out (see effective number of parameters in DIC)
  • 118. Effective dimension Understanding di in [A3]: In regular models, di dimension of T(Θi ), leading to BIC In non-regular models, di can be smaller
  • 119. Asymptotic marginals Asymptotically, under [A1]–[A4] mi (t) = gi (t|θi ) πi (θi ) dθi Θi is such that (i) if inf{|µi (θi ) − µ0 |; θi ∈ Θi } = 0, d−d Cl v n i mi (Tn ) d−d Cu vn i and (ii) if inf{|µi (θi ) − µ0 |; θi ∈ Θi } > 0 d−τ d−α mi (Tn ) = oPn [vn i + vn i ].
  • 120. Within-model consistency Under same assumptions, if inf{|µi (θi ) − µ0 |; θi ∈ Θi } = 0, the posterior distribution of µi (θi ) given Tn is consistent at rate 1/vn provided αi ∧ τi > di .
  • 121. Between-model consistency Consequence of above is that asymptotic behaviour of the Bayes factor is driven by the asymptotic mean value of Tn under both models. And only by this mean value!
  • 122. Between-model consistency Consequence of above is that asymptotic behaviour of the Bayes factor is driven by the asymptotic mean value of Tn under both models. And only by this mean value! Indeed, if inf{|µ0 − µ2 (θ2 )|; θ2 ∈ Θ2 } = inf{|µ0 − µ1 (θ1 )|; θ1 ∈ Θ1 } = 0 then −(d1 −d2 ) Cl vn m1 (Tn ) m2 (Tn ) −(d1 −d2 ) Cu vn , where Cl , Cu = OPn (1), irrespective of the true model. c Only depends on the difference d1 − d2 : no consistency
  • 123. Between-model consistency Consequence of above is that asymptotic behaviour of the Bayes factor is driven by the asymptotic mean value of Tn under both models. And only by this mean value! Else, if inf{|µ0 − µ2 (θ2 )|; θ2 ∈ Θ2 } > inf{|µ0 − µ1 (θ1 )|; θ1 ∈ Θ1 } = 0 then m1 (Tn ) m2 (Tn ) −(d1 −α2 ) Cu min vn −(d1 −τ2 ) , vn
  • 124. Consistency theorem If inf{|µ0 − µ2 (θ2 )|; θ2 ∈ Θ2 } = inf{|µ0 − µ1 (θ1 )|; θ1 ∈ Θ1 } = 0, Bayes factor −(d1 −d2 ) T B12 = O(vn ) irrespective of the true model. It is inconsistent since it always picks the model with the smallest dimension
  • 125. Consistency theorem If Pn belongs to one of the two models and if µ0 cannot be attained by the other one : 0 = min (inf{|µ0 − µi (θi )|; θi ∈ Θi }, i = 1, 2) < max (inf{|µ0 − µi (θi )|; θi ∈ Θi }, i = 1, 2) , T then the Bayes factor B12 is consistent
  • 126. Consequences on summary statistics Bayes factor driven by the means µi (θi ) and the relative position of µ0 wrt both sets {µi (θi ); θi ∈ Θi }, i = 1, 2. For ABC, this implies the simplest summary statistics Tn are ancillary statistics with different mean values under both models Else, if Tn asymptotically depends on some of the parameters of the models, it is possible that there exists θi ∈ Θi such that µi (θi ) = µ0 even though model M1 is misspecified
  • 127. Consequences on summary statistics Bayes factor driven by the means µi (θi ) and the relative position of µ0 wrt both sets {µi (θi ); θi ∈ Θi }, i = 1, 2. For ABC, this implies the simplest summary statistics Tn are ancillary statistics with different mean values under both models Else, if Tn asymptotically depends on some of the parameters of the models, it is possible that there exists θi ∈ Θi such that µi (θi ) = µ0 even though model M1 is misspecified
  • 128. Toy example: Laplace versus Gauss [1] If n Tn = n−1 Xi4 , µ1 (θ) = 3 + θ4 + 6θ2 , µ2 (θ) = 6 + · · · i=1 and the true distribution is Laplace with mean θ0 = 1, under the √ Gaussian model the value θ∗ = 2 3 − 3 also leads to µ0 = µ(θ∗ ) [here d1 = d2 = d = 1]
  • 129. Toy example: Laplace versus Gauss [1] If n n T =n Xi4 , −1 µ1 (θ) = 3 + θ4 + 6θ2 , µ2 (θ) = 6 + · · · i=1 and the true distribution is Laplace with mean θ0 = 1, under the √ Gaussian model the value θ∗ = 2 3 − 3 also leads to µ0 = µ(θ∗ ) [here d1 = d2 = d = 1] c A Bayes factor associated with such a statistic is inconsistent
  • 130. Toy example: Laplace versus Gauss [1] If n Tn = n−1 Xi4 , i=1 µ1 (θ) = 3 + θ4 + 6θ2 , µ2 (θ) = 6 + · · ·
  • 131. Toy example: Laplace versus Gauss [1] If n Tn = n−1 Xi4 , µ1 (θ) = 3 + θ4 + 6θ2 , i=1 q 0.35 0.40 0.45 n=1000 q q 0.30 q q M1 M2 Fourth moment µ2 (θ) = 6 + · · ·
  • 132. Toy example: Laplace versus Gauss [0] When (4) (6) ¯ ¯ T(y) = yn , yn and the true distribution is Laplace with mean θ0 = 0, then √ µ0 = 6, µ1 (θ∗ ) = 6 with θ∗ = 2 3 − 3 1 1 [d1 = 1 and d2 = 1/2] thus B12 ∼ n−1/4 → 0 : consistent Under the Gaussian model µ0 = 3, µ2 (θ2 ) B12 → +∞ : 6 > 3 ∀θ2 consistent
  • 133. Toy example: Laplace versus Gauss [0] When (4) (6) ¯ ¯ T(y) = yn , yn and the true distribution is Laplace with mean θ0 = 0, then √ µ0 = 6, µ1 (θ∗ ) = 6 with θ∗ = 2 3 − 3 1 1 [d1 = 1 and d2 = 1/2] thus B12 ∼ n−1/4 → 0 : consistent Under the Gaussian model µ0 = 3, µ2 (θ2 ) B12 → +∞ : 6 > 3 ∀θ2 consistent
  • 134. Toy example: Laplace versus Gauss [0] When (4) (6) ¯ ¯ T(y) = yn , yn 0.8 q q q q q 0.6 q 0.6 0.6 0.8 q n=10000 q q q q q q 1.0 q q 0.8 1.0 n=1000 1.0 n=100 q q Gauss Laplace 0.4 0.0 0.2 0.4 0.2 0.0 0.0 0.2 0.4 q q Gauss Laplace Fourth and sixth moments Gauss Laplace
  • 135. Checking for adequate statistics Run a practical check of the relevance (or non-relevance) of Tn null hypothesis that both models are compatible with the statistic Tn H0 : inf{|µ2 (θ2 ) − µ0 |; θ2 ∈ Θ2 } = 0 against H1 : inf{|µ2 (θ2 ) − µ0 |; θ2 ∈ Θ2 } > 0 testing procedure provides estimates of mean of Tn under each model and checks for equality
  • 136. Checking in practice Under each model Mi , generate ABC sample θi,l , l = 1, · · · , L For each θi,l , generate yi,l ∼ Fi,n (·|ψi,l ), derive Tn (yi,l ) and compute L 1 Tn (yi,l ), i = 1, 2 . µi = ^ L l=1 n Conditionally on T (y), √ L(µi − Eπ [µi (θi )|Tn (y)]) ^ N(0, Vi ), Test for a common mean H0 : µ1 ∼ N(µ0 , V1 ) , µ2 ∼ N(µ0 , V2 ) ^ ^ against the alternative of different means H1 : µi ∼ N(µi , Vi ), ^ with µ1 = µ2 .
  • 137. Toy example: Laplace versus Gauss q q 0 10 20 30 40 q q Gauss q q q Laplace Gauss Laplace Normalised χ2 without and with mad
  • 138. A population genetic illustration Two populations (1 and 2) having diverged at a fixed known time in the past and third population (3) which diverged from one of those two populations (models 1 and 2, respectively). Observation of 50 diploid individuals/population genotyped at 5, 50 or 100 independent microsatellite loci. Model 2
  • 139. A population genetic illustration Two populations (1 and 2) having diverged at a fixed known time in the past and third population (3) which diverged from one of those two populations (models 1 and 2, respectively). Observation of 50 diploid individuals/population genotyped at 5, 50 or 100 independent microsatellite loci. Stepwise mutation model: the number of repeats of the mutated gene increases or decreases by one. Mutation rate µ common to all loci set to 0.005 (single parameter) with uniform prior distribution µ ∼ U[0.0001, 0.01]
  • 140. A population genetic illustration Summary statistics associated to the (δµ )2 distance xl,i,j repeated number of allele in locus l = 1, . . . , L for individual i = 1, . . . , 100 within the population j = 1, 2, 3. Then (δµ )21 ,j2 = j 1 L L l=1   1 100 100 xl,i1 ,j1 − i1 =1 1 100 100 2 xl,i2 ,j2  . i2 =1
  • 141. A population genetic illustration For two copies of locus l with allele sizes xl,i,j1 and xl,i ,j2 , most recent common ancestor at coalescence time τj1 ,j2 , gene genealogy distance of 2τj1 ,j2 , hence number of mutations Poisson with parameter 2µτj1 ,j2 . Therefore, xl,i,j1 − xl,i E 2 ,j2 |τj1 ,j2 = 2µτj1 ,j2 and Model 1 Model 2 (δµ )2 1,2 2µ1 t 2µ2 t E (δµ )2 1,3 2µ1 t 2µ2 t E (δµ )2 2,3 2µ1 t 2µ2 t E
  • 142. A population genetic illustration Thus, Bayes factor based only on distance (δµ )2 not convergent: if 1,2 µ1 = µ2 , same expectation Bayes factor based only on distance (δµ )2 or (δµ )2 not 1,3 2,3 convergent: if µ1 = 2µ2 or 2µ1 = µ2 same expectation if two of the three distances are used, Bayes factor converges: there is no (µ1 , µ2 ) for which all expectations are equal
  • 143. A population genetic illustration 0.8 DM2(12) q 50 100 0.0 0.4 q q 5 0.8 DM2(13) 0.4 q q q q q q q q q q q 0.0 q 5 50 100 DM2(13) & DM2(23) 0.8 q q q q q q q q q q q q q q 0.4 q q q q 0.0 q q q q q 5 50 100 Posterior probabilities that the data is from model 1 for 5, 50 and 100 loci
  • 144. A population genetic illustration 20 0.2 40 0.4 60 80 0.6 100 0.8 120 1.0 140 p−values 0 0.0 q q q DM2(12) DM2(13) & DM2(23) q DM2(12) DM2(13) & DM2(23)
  • 145. Embedded models When M1 submodel of M2 , and if the true distribution belongs to the smaller model M1 , Bayes factor is of order −(d1 −d2 ) vn
  • 146. Embedded models When M1 submodel of M2 , and if the true distribution belongs to the smaller model M1 , Bayes factor is of order −(d1 −d2 ) vn If summary statistic only informative on a parameter that is the same under both models, i.e if d1 = d2 , then c the Bayes factor is not consistent
  • 147. Embedded models When M1 submodel of M2 , and if the true distribution belongs to the smaller model M1 , Bayes factor is of order −(d1 −d2 ) vn Else, d1 < d2 and Bayes factor is going to ∞ under M1 . If true distribution not in M1 , then c Bayes factor is consistent only if µ1 = µ2 = µ0
  • 148. Summary Model selection feasible with ABC Choice of summary statistics is paramount At best, ABC → π(. | T(y)) which concentrates around µ0 For consistent estimation : {θ; µ(θ) = µ0 } = θ0 For consistent testing {µ1 (θ1 ), θ1 ∈ Θ1 } ∩ {µ2 (θ2 ), θ2 ∈ Θ2 } = ∅
  • 149. Summary Model selection feasible with ABC Choice of summary statistics is paramount At best, ABC → π(. | T(y)) which concentrates around µ0 For consistent estimation : {θ; µ(θ) = µ0 } = θ0 For consistent testing {µ1 (θ1 ), θ1 ∈ Θ1 } ∩ {µ2 (θ2 ), θ2 ∈ Θ2 } = ∅
  • 150. Conclusion/perspectives abc part of a wider picture to handle complex/Big Data models, able to start from rudimentary machine learning summaries many formats of empirical [likelihood] Bayes methods available lack of comparative tools and of an assessment for information loss