Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

31 more user friendly things #36

Merged
merged 3 commits into from
Jul 16, 2024
Merged

31 more user friendly things #36

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
520dc87
do not resample hyper when Q is invalid #31
adamwang15 Jul 16, 2024
b26a265
update readme example
adamwang15 Jul 16, 2024
b140100
comment about notations
adamwang15 Jul 16, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
51 changes: 23 additions & 28 deletions README.Rmd
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,17 +29,35 @@ devtools::install_github("bsvars/bsvarSIGNs")

# Example

A repliaction of Antolín-Díaz and Rubio-Ramírez (2018).
A replication of Arias, Rubio-Ramírez and Waggoner (2018).

```r
data(optimism)

# optimism shock
# no effect on productivity + positive effect on stock prices
sign_irf = matrix(c(0, 1, rep(NA, 23)), 5, 5)

specification = specify_bsvarSIGN$new(optimism * 100,
p = 4,
sign_irf = sign_irf)

posterior = estimate(specification, S = 100)
irf = compute_impulse_responses(posterior, horizon = 40)
plot(irf, probability = 0.68)
```

A replication of Antolín-Díaz and Rubio-Ramírez (2018).

```r
data(monetary)

# contractionary monetary policy shock
sign_irf = matrix(0, 6, 6)
sign_irf[, 1] = c(0, -1, -1, 0, -1, 1)
sign_irf = array(sign_irf, dim = c(6, 6, 5))
sign_irf = matrix(NA, 6, 6)
sign_irf[, 1] = c(NA, -1, -1, NA, -1, 1)
sign_irf = array(sign_irf, dim = c(6, 6, 6))

# in October 1979 the shock
# in October 1979
sign_narrative = rbind(c(1, 1, NA, 1, 166, 0), # is positive
c(3, 1, 6, 1, 166, 0)) # greatest historical decomposition

Expand All @@ -53,26 +71,3 @@ irf = compute_impulse_responses(posterior, horizon = 60)
plot(irf, probability = 0.68)
```

A repliaction of Arias, Rubio-Ramírez and Waggoner (2018).

```r
data(optimism)

# optimism shock
# no contemporaneous effect on productivity
zero_irf = matrix(0, nrow = 5, ncol = 5)
zero_irf[1, 1] = 1
# positive contemporaneous effect on stock prices
sign_irf = array(0, dim = c(5, 5, 1))
sign_irf[2, 1, 1] = 1

specification = specify_bsvarSIGN$new(optimism * 100,
p = 4,
sign_irf = sign_irf,
zero_irf = zero_irf)

posterior = estimate(specification, S = 100)
irf = compute_impulse_responses(posterior, horizon = 40)
plot(irf, probability = 0.68)
```

51 changes: 23 additions & 28 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,17 +18,35 @@ devtools::install_github("bsvars/bsvarSIGNs")

# Example

A repliaction of Antolín-Díaz and Rubio-Ramírez (2018).
A replication of Arias, Rubio-Ramírez and Waggoner (2018).

``` r
data(optimism)

# optimism shock
# no effect on productivity + positive effect on stock prices
sign_irf = matrix(c(0, 1, rep(NA, 23)), 5, 5)

specification = specify_bsvarSIGN$new(optimism * 100,
p = 4,
sign_irf = sign_irf)

posterior = estimate(specification, S = 100)
irf = compute_impulse_responses(posterior, horizon = 40)
plot(irf, probability = 0.68)
```

A replication of Antolín-Díaz and Rubio-Ramírez (2018).

``` r
data(monetary)

# contractionary monetary policy shock
sign_irf = matrix(0, 6, 6)
sign_irf[, 1] = c(0, -1, -1, 0, -1, 1)
sign_irf = array(sign_irf, dim = c(6, 6, 5))
sign_irf = matrix(NA, 6, 6)
sign_irf[, 1] = c(NA, -1, -1, NA, -1, 1)
sign_irf = array(sign_irf, dim = c(6, 6, 6))

# in October 1979 the shock
# in October 1979
sign_narrative = rbind(c(1, 1, NA, 1, 166, 0), # is positive
c(3, 1, 6, 1, 166, 0)) # greatest historical decomposition

Expand All @@ -41,26 +59,3 @@ posterior = estimate(specification, S = 100)
irf = compute_impulse_responses(posterior, horizon = 60)
plot(irf, probability = 0.68)
```

A repliaction of Arias, Rubio-Ramírez and Waggoner (2018).

``` r
data(optimism)

# optimism shock
# no contemporaneous effect on productivity
zero_irf = matrix(0, nrow = 5, ncol = 5)
zero_irf[1, 1] = 1
# positive contemporaneous effect on stock prices
sign_irf = array(0, dim = c(5, 5, 1))
sign_irf[2, 1, 1] = 1

specification = specify_bsvarSIGN$new(optimism * 100,
p = 4,
sign_irf = sign_irf,
zero_irf = zero_irf)

posterior = estimate(specification, S = 100)
irf = compute_impulse_responses(posterior, horizon = 40)
plot(irf, probability = 0.68)
```
65 changes: 37 additions & 28 deletions src/bsvars_sign.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,14 @@ using namespace Rcpp;
using namespace arma;


/***
All notations in the C++ code except for compute.cpp and forecast_bsvarSIGNs.cpp
are consistent with the notations in the papers:
Antolín-Díaz and Rubio-Ramírez (2018) and Arias, Rubio-Ramírez and Waggoner (2018)
which are different from the notations in the R code.
***/


// [[Rcpp::interfaces(cpp)]]
// [[Rcpp::export]]
Rcpp::List bsvar_sign_cpp(
Expand Down Expand Up @@ -47,7 +55,7 @@ Rcpp::List bsvar_sign_cpp(
if (show_progress) {
Rcout << "**************************************************|" << endl;
Rcout << " bsvarSIGNs: Bayesian Structural VAR with zero, |" << endl;
Rcout << " sign and narrative restrictions |" << endl;
Rcout << " sign and narrative restrictions |" << endl;
Rcout << "**************************************************|" << endl;
// Rcout << " Gibbs sampler for the SVAR model |" << endl;
// Rcout << "**************************************************|" << endl;
Expand Down Expand Up @@ -96,32 +104,36 @@ Rcpp::List bsvar_sign_cpp(
// #pragma omp parallel for private(hyper, mu, delta, lambda, psi, prior_V, prior_S, Ystar, Xstar, Yplus, Xplus, result, post_B, post_V, post_S, Sigma, chol_Sigma, B, h_invp, Q, shocks, w)
for (int s = 0; s < S; s++) {

w = 0;
hyper = hypers.col(randi(distr_param(0, S_hyper)));
mu = hyper(0);
delta = hyper(1);
lambda = hyper(2);
psi = hyper.rows(3, N + 2);

// update Minnesota prior
prior_V = diagmat(prior_v %
join_vert(lambda * lambda * repmat(1 / psi, p, 1),
ones<vec>(K - N * p)));
prior_S = diagmat(psi);

// update dummy observation prior
Ystar = join_vert(Ysoc / mu, Ysur / delta);
Xstar = join_vert(Xsoc / mu, Xsur / delta);
Yplus = join_vert(Ystar, Y);
Xplus = join_vert(Xstar, X);

// posterior parameters
result = niw_cpp(Yplus, Xplus, prior_B, prior_V, prior_S, prior_nu);
post_B = result(0);
post_V = result(1);
post_S = result(2);
post_nu = as_scalar(result(3));

w = 0;

while (w == 0) {
hyper = hypers.col(randi(distr_param(0, S_hyper)));
mu = hyper(0);
delta = hyper(1);
lambda = hyper(2);
psi = hyper.rows(3, N + 2);

// update Minnesota prior
prior_V = diagmat(prior_v %
join_vert(lambda * lambda * repmat(1 / psi, p, 1),
ones<vec>(K - N * p)));
prior_S = diagmat(psi);

// update dummy observation prior
Ystar = join_vert(Ysoc / mu, Ysur / delta);
Xstar = join_vert(Xsoc / mu, Xsur / delta);
Yplus = join_vert(Ystar, Y);
Xplus = join_vert(Xstar, X);

// posterior parameters
result = niw_cpp(Yplus, Xplus, prior_B, prior_V, prior_S, prior_nu);
post_B = result(0);
post_V = result(1);
post_S = result(2);
post_nu = as_scalar(result(3));
checkUserInterrupt();

// sample reduced-form parameters
Sigma = iwishrnd(post_S, post_nu);
Expand All @@ -145,9 +157,6 @@ Rcpp::List bsvar_sign_cpp(
posterior_Theta0.slice(s) = chol_Sigma * Q;
posterior_shocks.slice(s) = shocks;

// Check for user interrupts
if (s % 200 == 0) checkUserInterrupt();

// Increment progress bar
if (any(prog_rep_points == s)) bar.increment();

Expand Down
2 changes: 1 addition & 1 deletion src/restrictions_narrative.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -79,7 +79,7 @@ double weight_narrative(
const arma::cube& irf
) {

const int M = 1e+04; // number of draws to approximate normal distribution
const int M = 1e+3; // number of draws to approximate normal distribution

double n_success = 1.0e-15;

Expand Down
Loading