sequential monte carlo methods for dsge models 1 · sequential monte carlo methods for dsge models...
TRANSCRIPT
![Page 1: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/1.jpg)
Sequential Monte Carlo Methods for DSGE Models 1
Ed Herbst∗ Frank Schorfheide+
∗Federal Reserve Board
+University of Pennsylvania, PIER, CEPR, and NBER
June 12, 2017
1The views expressed in this presentation are those of the presenters and do not necessarily reflect the viewsof the Federal Reserve Board of Governors or the Federal Reserve System.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 2: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/2.jpg)
Some References
These lectures use material from our joint work:
• “Tempered Particle Filtering,” 2016, PIER Working Paper, 16-017
• Bayesian Estimation of DSGE Models, 2015, Princeton University Press
• “Sequential Monte Carlo Sampling for DSGE Models,” 2014, Journal of Econometrics
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 3: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/3.jpg)
Some Background
• DSGE model: dynamic model of the macroeconomy, indexed by θ – vector of preferenceand technology parameters. Used for forecasting, policy experiments, interpreting pastevents.
• Bayesian analysis of DSGE models:
p(θ|Y ) =p(Y |θ)p(θ)
p(Y )∝ p(Y |θ)p(θ).
• Computational hurdles: numerical solution of model leads to state-space representation=⇒ likelihood approximation =⇒ posterior sampler.
• “Standard” approach for (linearized) models (Schorfheide, 2000; Otrok, 2001):• Model solution: log-linearize and use linear rational expectations system solver.• Evaluation of p(Y |θ): Kalman filter• Posterior draws θi : MCMC
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 4: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/4.jpg)
Sequential Monte Carlo (SMC) Methods
SMC can help to
Lecture 1
• approximate the posterior of θ: Chopin (2002) ... Durham and Geweke (2013) ... Creal(2007), Herbst and Schorfheide (2014)
Lecture 2
• approximate the likelihood function (particle filtering): Gordon, Salmond, and Smith(1993) ... Fernandez-Villaverde and Rubio-Ramirez (2007)
• or both: SMC 2: Chopin, Jacob, and Papaspiliopoulos (2012) ... Herbst and Schorfheide(2015)
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 5: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/5.jpg)
Lecture 1
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 6: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/6.jpg)
Sampling from Posterior
• DSGE model posteriors are often non-elliptical, e.g., multimodal posteriors may arise
because it is difficult to
• disentangle internaland externalpropagationmechanisms;
• disentangle therelative importance ofshocks.
0.0 0.2 0.4 0.6 0.8 1.00.0
0.2
0.4
0.6
0.8
1.0θ2
θ1
• Economic Example: is wage growth persistent because
1 wage setters find it very costly to adjust wages?
2 exogenous shocks affect the substitutability of labor inputs and hence markups?
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 7: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/7.jpg)
Sampling from Posterior
• If posterior distributions are irregular, standard MCMC methods can be inaccurate(examples will follow).
• SMC samplers often generate more precise approximations of posteriors in the sameamount of time.
• SMC can be parallelized.
• SMC = importance sampling on steroids =⇒ We will first review importance sampling.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 8: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/8.jpg)
Importance Sampling
• Approximate π(·) by using a different, tractable density g(θ) that is easy to sample from.
• For more general problems, posterior density may be unnormalized. So we write
π(θ) =p(Y |θ)p(θ)
p(Y )=
f (θ)∫f (θ)dθ
.
• Importance sampling is based on the identity
Eπ[h(θ)] =
∫h(θ)π(θ)dθ =
∫Θh(θ) f (θ)
g(θ)g(θ)dθ∫Θ
f (θ)g(θ)g(θ)dθ
.
• (Unnormalized) importance weight:
w(θ) =f (θ)
g(θ).
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 9: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/9.jpg)
Importance Sampling
1 For i = 1 to N, draw θi iid∼ g(θ) and compute the unnormalized importance weights
w i = w(θi ) =f (θi )
g(θi ).
2 Compute the normalized importance weights
W i =w i
1N
∑Ni=1 w
i.
An approximation of Eπ[h(θ)] is given by
hN =1
N
N∑i=1
W ih(θi ).
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 10: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/10.jpg)
Illustration
If θi ’s are draws from g(·) then
Eπ[h] ≈1N
∑Ni=1 h(θi )w(θi )
1N
∑Ni=1 w(θi )
, w(θ) =f (θ)
g(θ).
6 4 2 0 2 4 60.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
f
g1
g2
6 4 2 0 2 4 6
weights
f/g1
f/g2
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 11: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/11.jpg)
Accuracy
• Since we are generating iid draws from g(θ), it’s fairly straightforward to derive a CLT:
√N(hN − Eπ[h]) =⇒ N
(0,Ω(h)
), where Ω(h) = Vg [(π/g)(h − Eπ[h])].
• Using a crude approximation (see, e.g., Liu (2008)), we can factorize Ω(h) as follows:
Ω(h) ≈ Vπ[h](Vg [π/g ] + 1
).
The approximation highlights that the larger the variance of the importance weights, theless accurate the Monte Carlo approximation relative to the accuracy that could beachieved with an iid sample from the posterior.
• Users often monitor
ESS = NVπ[h]
Ω(h)≈ N
1 + Vg [π/g ].
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 12: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/12.jpg)
From Importance Sampling to Sequential Importance Sampling
• In general, it’s hard to construct a good proposal density g(θ),
• especially if the posterior has several peaks and valleys.
• Idea - Part 1: it might be easier to find a proposal density for
πn(θ) =[p(Y |θ)]φnp(θ)∫[p(Y |θ)]φnp(θ)dθ
=fn(θ)
Zn.
at least if φn is close to zero.
• Idea - Part 2: We can try to turn a proposal density for πn into a proposal density for πn+1
and iterate, letting φn −→ φN = 1.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 13: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/13.jpg)
Illustration: Tempered Posteriors of θ1
θ1
0.00.2
0.40.6
0.81.0
n10
2030
4050
0
1
2
3
4
5
πn(θ) =[p(Y |θ)]φnp(θ)∫[p(Y |θ)]φnp(θ)dθ
=fn(θ)
Zn, φn =
(n
Nφ
)λE. Herbst and F. Schorfheide SMC for DSGE Models
![Page 14: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/14.jpg)
SMC Algorithm: A Graphical Illustration
C S M C S M C S M
−10
−5
0
5
10
φ0 φ1 φ2 φ3
• πn(θ) is represented by a swarm of particles θin,W
inN
i=1:
hn,N =1
N
N∑i=1
W inh(θi
n)a.s.−→ Eπn [h(θn)].
• C is Correction; S is Selection; and M is Mutation.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 15: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/15.jpg)
SMC Algorithm
1 Initialization. (φ0 = 0). Draw the initial particles from the prior: θi1
iid∼ p(θ) and W i1 = 1,
i = 1, . . . ,N.
2 Recursion. For n = 1, . . . ,Nφ,
1 Correction. Reweight the particles from stage n − 1 by defining the incremental weights
w in = [p(Y |θi
n−1)]φn−φn−1 (1)
and the normalized weights
W in =
w inW
in−1
1N
∑Ni=1 w
inW i
n−1
, i = 1, . . . ,N. (2)
An approximation of Eπn [h(θ)] is given by
hn,N =1
N
N∑i=1
W inh(θi
n−1). (3)
2 Selection.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 16: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/16.jpg)
SMC Algorithm
1 Initialization.
2 Recursion. For n = 1, . . . ,Nφ,
1 Correction.2 Selection. (Optional Resampling) Let θN
i=1 denote N iid draws from a multinomialdistribution characterized by support points and weights θi
n−1, WinN
i=1 and set W in = 1.
An approximation of Eπn [h(θ)] is given by
hn,N =1
N
N∑i=1
W inh(θi
n). (4)
3 Mutation. Propagate the particles θi ,Win via NMH steps of a MH algorithm with
transition density θin ∼ Kn(θn|θi
n; ζn) and stationary distribution πn(θ). An approximation ofEπn [h(θ)] is given by
hn,N =1
N
N∑i=1
h(θin)W i
n . (5)
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 17: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/17.jpg)
Remarks
• Correction Step:• reweight particles from iteration n − 1 to create importance sampling approximation of
Eπn [h(θ)]
• Selection Step: the resampling of the particles• (good) equalizes the particle weights and thereby increases accuracy of subsequent
importance sampling approximations;• (not good) adds a bit of noise to the MC approximation.
• Mutation Step: changes particle values• adapts particles to posterior πn(θ);• imagine we don’t do it: then we would be using draws from prior p(θ) to approximate
posterior π(θ), which can’t be good!
θ1
0.00.2
0.40.6
0.81.0
n10
2030
4050
0
1
2
3
4
5
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 18: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/18.jpg)
More on Transition Kernel in Mutation Step
• Transition kernel Kn(θ|θn−1; ζn): generated by running M steps of a Metropolis-Hastingsalgorithm.
• Lessons from DSGE model MCMC:• blocking of parameters can reduces persistence of Markov chain;• mixture proposal density avoids “getting stuck.”
• Blocking: Partition the parameter vector θn into Nblocks equally sized blocks, denoted byθn,b, b = 1, . . . ,Nblocks . (We generate the blocks for n = 1, . . . ,Nφ randomly prior torunning the SMC algorithm.)
• Example: random walk proposal density:
ϑb|(θin,b,m−1, θ
in,−b,m,Σ
∗n,b) ∼ N
(θi
n,b,m−1, c2n Σ∗
n,b
).
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 19: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/19.jpg)
Adaptive Choice of ζn = (Σ∗n, cn)
• Infeasible adaption:• Let Σ∗
n = Vπn [θ].• Adjust scaling factor according to
cn = cn−1f(1− Rn−1(ζn−1)
),
where Rn−1(·) is population rejection rate from iteration n − 1 and
f (x) = 0.95 + 0.10e16(x−0.25)
1 + e16(x−0.25).
• Feasible adaption – use output from stage n − 1 to replace ζn by ζn:
• Use particle approximations of Eπn [θ] and Vπn [θ] based on θin−1, W
inN
i=1.
• Use actual rejection rate from stage n − 1 to calculate cn = cn−1f(Rn−1(ζn−1)
).
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 20: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/20.jpg)
More on Resampling
• So far, we have used multinomial resampling. It’s fairly intuitive and it is straightforwardto obtain a CLT.
• But: multinominal resampling is not particularly efficient.
• The Herbst-Schorfheide book contains a section on alternative resampling schemes(stratified resampling, residual resampling...)
• These alternative techniques are designed to achieve a variance reduction.
• Most resampling algorithms are not parallelizable because they rely on the normalizedparticle weights.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 21: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/21.jpg)
Application 1: Small Scale New Keynesian Model
• We will take a look at the effect of various tuning choices on accuracy:
• Tempering schedule λ: λ = 1 is linear, λ > 1 is convex.
• Number of stages Nφ versus number of particles N.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 22: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/22.jpg)
Effect of λ on Inefficiency Factors InEffN [θ]
1 2 3 4 5 6 7 8100
101
102
103
104
λ
Notes: The figure depicts hairs of InEffN [θ] as function of λ. The inefficiency factors arecomputed based on Nrun = 50 runs of the SMC algorithm. Each hair corresponds to a DSGEmodel parameter.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 23: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/23.jpg)
Number of Stages Nφ vs Number of Particles N
ρg σr ρr ρz σg κ σz π(A) ψ1 γ(Q) ψ2 r (A) τ10−3
10−2
10−1
100
101
Nφ = 400, N = 250Nφ = 200, N = 500Nφ = 100, N = 1000
Nφ = 50, N = 2000Nφ = 25, N = 4000
Notes: Plot of V[θ]/Vπ[θ] for a specific configuration of the SMC algorithm. The inefficiencyfactors are computed based on Nrun = 50 runs of the SMC algorithm. Nblocks = 1, λ = 2,NMH = 1.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 24: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/24.jpg)
A Few Words on Posterior Model Probabilities
• Posterior model probabilities
πi,T =πi,0p(Y1:T |Mi )∑M
j=1 πj,0p(Y1:T |Mj )
where
p(Y1:T |Mi ) =
∫p(Y1:T |θ(i),Mi )p(θ(i)|Mi )dθ(i)
• For any model:
ln p(Y1:T |Mi ) =T∑
t=1
ln
∫p(yt |θ(i),Y1:t−1,Mi )p(θ(i)|Y1:t−1,Mi )dθ(i)
• Marginal data density p(Y1:T |Mi ) arises as a by-product of SMC.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 25: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/25.jpg)
Marginal Likelihood Approximation
• Recall w in = [p(Y |θi
n−1)]φn−φn−1 .
• Then
1
N
N∑i=1
w inW
in−1 ≈
∫[p(Y |θ)]φn−φn−1
pφn−1 (Y |θ)p(θ)∫pφn−1 (Y |θ)p(θ)dθ
dθ
=
∫p(Y |θ)φnp(θ)dθ∫p(Y |θ)φn−1p(θ)dθ
• Thus,
Nφ∏n=1
(1
N
N∑i=1
w inW
in−1
)≈∫
p(Y |θ)p(θ)dθ.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 26: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/26.jpg)
SMC Marginal Data Density Estimates
Nφ = 100 Nφ = 400N Mean(ln p(Y )) SD(ln p(Y )) Mean(ln p(Y )) SD(ln p(Y ))500 -352.19 (3.18) -346.12 (0.20)1,000 -349.19 (1.98) -346.17 (0.14)2,000 -348.57 (1.65) -346.16 (0.12)4,000 -347.74 (0.92) -346.16 (0.07)
Notes: Table shows mean and standard deviation of log marginal data density estimates as afunction of the number of particles N computed over Nrun = 50 runs of the SMC sampler withNblocks = 4, λ = 2, and NMH = 1.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 27: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/27.jpg)
Application 2: Estimation of Smets and Wouters (2007) Model
• Benchmark macro model, has been estimated many (many) times.
• “Core” of many larger-scale models.
• 36 estimated parameters.
• RWMH: 10 million draws (5 million discarded); SMC: 500 stages with 12,000 particles.
• We run the RWM (using a particular version of a parallelized MCMC) and the SMCalgorithm on 24 processors for the same amount of time.
• We estimate the SW model twenty times using RWM and SMC and get essentiallyidentical results.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 28: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/28.jpg)
Application 2: Estimation of Smets and Wouters (2007) Model
• More interesting question: how does quality of posterior simulators change as one makesthe priors more diffuse?
• Replace Beta by Uniform distributions; increase variances of parameters with Gamma andNormal prior by factor of 3.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 29: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/29.jpg)
SW Model with DIFFUSE Prior: Estimation stability RWH (black) versusSMC (red)
l ιw µp µw ρw ξw rπ−4
−3
−2
−1
0
1
2
3
4
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 30: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/30.jpg)
A Measure of Effective Number of Draws
• Suppose we could generate iid Neff draws from posterior, then
Eπ[θ]approx∼ N
(Eπ[θ],
1
NeffVπ[θ]
).
• We can measure the variance of Eπ[θ] by running SMC and RWM algorithm repeatedly.
• Then,
Neff ≈Vπ[θ]
V[Eπ[θ]
]
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 31: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/31.jpg)
Effective Number of Draws
SMC RWMHParameter Mean STD(Mean) Neff Mean STD(Mean) Neff
σl 3.06 0.04 1058 3.04 0.15 60l -0.06 0.07 732 -0.01 0.16 177ιp 0.11 0.00 637 0.12 0.02 19h 0.70 0.00 522 0.69 0.03 5Φ 1.71 0.01 514 1.69 0.04 10rπ 2.78 0.02 507 2.76 0.03 159ρb 0.19 0.01 440 0.21 0.08 3ϕ 8.12 0.16 266 7.98 1.03 6σp 0.14 0.00 126 0.15 0.04 1ξp 0.72 0.01 91 0.73 0.03 5ιw 0.73 0.02 87 0.72 0.03 36µp 0.77 0.02 77 0.80 0.10 3ρw 0.69 0.04 49 0.69 0.09 11µw 0.63 0.05 49 0.63 0.09 11ξw 0.93 0.01 43 0.93 0.02 8
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 32: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/32.jpg)
A Closer Look at the Posterior: Two Modes
Parameter Mode 1 Mode 2ξw 0.844 0.962ιw 0.812 0.918ρw 0.997 0.394µw 0.978 0.267Log Posterior -804.14 -803.51
• Mode 1 implies that wage persistence is driven by extremely exogenous persistent wagemarkup shocks.
• Mode 2 implies that wage persistence is driven by endogenous amplification of shocksthrough the wage Calvo and indexation parameter.
• SMC is able to capture the two modes.
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 33: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/33.jpg)
A Closer Look at the Posterior: Internal ξw versus External ρw Propagation
0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 10
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
ξw
ρw
E. Herbst and F. Schorfheide SMC for DSGE Models
![Page 34: Sequential Monte Carlo Methods for DSGE Models 1 · Sequential Monte Carlo Methods for DSGE Models 1 Ed Herbst Frank Schorfheide+ Federal Reserve Board +University of Pennsylvania,](https://reader036.vdocument.in/reader036/viewer/2022071009/5fc6a034d4879e2f0b063d7e/html5/thumbnails/34.jpg)
Stability of Posterior Computations: RWH (black) versus SMC (red)
P (ξw > ρw) P (ρw > µw) P (ξw > µw) P (ξp > ρp) P (ρp > µp) P (ξp > µp)0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
E. Herbst and F. Schorfheide SMC for DSGE Models