Compute linear impulse responses with identified shock and/or with 2SLS
Compute linear impulse responses with identified shock and/or with 2SLS.
lp_lin_iv( endog_data, shock = NULL, instr = NULL, use_twosls = FALSE, instrum = NULL, lags_endog_lin = NULL, exog_data = NULL, lags_exog = NULL, contemp_data = NULL, lags_criterion = NaN, max_lags = NaN, trend = NULL, confint = NULL, use_nw = TRUE, nw_lag = NULL, nw_prewhite = FALSE, adjust_se = FALSE, hor = NULL, num_cores = 1 )
endog_data |
A data.frame, containing the values of the dependent variable(s). |
shock |
A one column data.frame, including the variable to shock with. The row length has to be the same as endog_data. When use_twosls = TRUE, this variable will be approximated/regressed on the instrument variable(s) given in instrum. |
instr |
Deprecated input name. Use shock instead. See shock for details. |
use_twosls |
Boolean. Use two stage least squares? TRUE or FALSE (default). |
instrum |
A data.frame, containing the instrument(s) to use for 2SLS. This instrument will be used for the variable in shock. |
lags_endog_lin |
NaN or integer. NaN if lags are chosen by a lag length criterion. Integer for number of lags for endog_data. |
exog_data |
A data.frame, containing exogenous variables. The row length has to be the same as endog_data. Lag lengths for exogenous variables have to be given and will no be determined via a lag length criterion. |
lags_exog |
NULL or Integer. Integer for the number of lags for the exogenous data. |
contemp_data |
A data.frame, containing exogenous data with contemporaneous impact. The row length has to be the same as endog_data. |
lags_criterion |
NaN or character. NaN means that the number of lags will be given at lags_endog_lin. Possible lag length criteria are 'AICc', 'AIC' or 'BIC'. Note that when use_twosls = TRUE, the lag lengths are chosen based on normal OLS regressions, without using the instruments. |
max_lags |
NaN or integer. Maximum number of lags if lags_criterion is a character denoting the lag length criterion. NaN otherwise. |
trend |
Integer. No trend = 0 , include trend = 1, include trend and quadratic trend = 2. |
confint |
Double. Width of confidence bands. 68% = 1; 90% = 1.65; 95% = 1.96. |
use_nw |
Boolean. Use Newey-West (1987) standard errors for impulse responses? TRUE (default) or FALSE. |
nw_lag |
Integer. Specifies the maximum lag with positive weight for the Newey-West estimator. If set to NULL (default), the lag increases with with the number of horizon. |
nw_prewhite |
Boolean. Should the estimators be pre-whitened? TRUE of FALSE (default). |
adjust_se |
Boolen. Should a finite sample adjsutment be made to the covariance matrix estimators? TRUE or FALSE (default). |
hor |
Integer. Number of horizons for impulse responses. |
num_cores |
NULL or Integer. The number of cores to use for the estimation. If NULL, the function will use the maximum number of cores minus one. |
A list containing:
irf_lin_mean |
A matrix, containing the impulse responses. The row in each matrix denotes the response of the ith variable to the shock. The columns are the horizons. |
irf_lin_low |
A matrix, containing all lower confidence bands of the impulse responses, based on robust standard errors by Newey and West (1987). Properties are equal to irf_lin_mean. |
irf_lin_up |
A matrix, containing all upper confidence bands of the impulse responses, based on robust standard errors by Newey and West (1987). Properties are equal to irf_lin_mean. |
specs |
A list with properties of endog_data for the plot function. It also contains lagged data (y_lin and x_lin) used for the estimations of the impulse responses, and the selected lag lengths when an information criterion has been used. |
Philipp Adämmer
Akaike, H. (1974). "A new look at the statistical model identification", IEEE Transactions on Automatic Control, 19 (6): 716–723.
Auerbach, A. J., and Gorodnichenko, Y. (2012). "Measuring the Output Responses to Fiscal Policy." American Economic Journal: Economic Policy, 4 (2): 1-27.
Blanchard, O., and Perotti, R. (2002). “An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output.” Quarterly Journal of Economics, 117(4): 1329–1368.
Hurvich, C. M., and Tsai, C.-L. (1989), "Regression and time series model selection in small samples", Biometrika, 76(2): 297–307
Jordà, Ò. (2005). "Estimation and Inference of Impulse Responses by Local Projections." American Economic Review, 95 (1): 161-182.
Jordà, Ò, Schularick, M., Taylor, A.M. (2015), "Betting the house", Journal of International Economics, 96, S2-S18.
Newey, W.K., and West, K.D. (1987). “A Simple, Positive-Definite, Heteroskedasticity and Autocorrelation Consistent Covariance Matrix.” Econometrica, 55: 703–708.
Ramey, V.A., and Zubairy, S. (2018). "Government Spending Multipliers in Good Times and in Bad: Evidence from US Historical Data." Journal of Political Economy, 126(2): 850 - 901.
Schwarz, Gideon E. (1978). "Estimating the dimension of a model", Annals of Statistics, 6 (2): 461–464.
# This example replicates a result from the Supplementary Appendix
# by Ramey and Zubairy (2018) (RZ-18)
# Load data
ag_data <- ag_data
sample_start <- 7
sample_end <- dim(ag_data)[1]
# Endogenous data
endog_data <- ag_data[sample_start:sample_end,3:5]
# Variable to shock with. Here government spending due to
# Blanchard and Perotti (2002) framework
shock <- ag_data[sample_start:sample_end, 3]
# Estimate linear model
results_lin_iv <- lp_lin_iv(endog_data,
lags_endog_lin = 4,
shock = shock,
trend = 0,
confint = 1.96,
hor = 20)
# Show all impulse responses
plot(results_lin_iv)
# Make and save plots
iv_lin_plots <- plot_lin(results_lin_iv)
# * The first element of 'iv_lin_plots' shows the response of the first
# variable (Gov) to the shock (Gov).
# * The second element of 'iv_lin_plots' shows the response of the second
# variable (Tax) to the shock (Gov).
# * ...
# This plot replicates the left plot in the mid-panel of Figure 12 in the
# Supplementary Appendix by RZ-18.
iv_lin_plots[[1]]
# Show diagnostics. The first element shows the reaction of the first given endogenous variable.
summary(results_lin_iv)
## Add lags of the identified shock ##
# Endogenous data but now exclude government spending
endog_data <- ag_data[sample_start:sample_end, 4:5]
# Variable to shock with (government spending)
shock <- ag_data[sample_start:sample_end, 3]
# Add the shock variable to exogenous data
exog_data <- shock
# Estimate linear model with lagged shock variable
results_lin_iv <- lp_lin_iv(endog_data,
lags_endog_lin = 4,
shock = shock,
exog_data = exog_data,
lags_exog = 2,
trend = 0,
confint = 1.96,
hor = 20)
# Show all responses
plot(results_lin_iv)
# Show diagnostics. The first element shows the reaction of the first endogenous variable.
summary(results_lin_iv)
##############################################################################
##### Use 2SLS #########
##############################################################################
# Set seed
set.seed(007)
# Load data
ag_data <- ag_data
sample_start <- 7
sample_end <- dim(ag_data)[1]
# Endogenous data
endog_data <- ag_data[sample_start:sample_end,3:5]
# Variable to shock with (government spending)
shock <- ag_data[sample_start:sample_end, 3]
# Generate instrument variable that is correlated with government spending
instrum <- as.data.frame(0.9*shock$Gov + rnorm(length(shock$Gov), 0, 0.02) )
# Estimate linear model via 2SLS
results_lin_iv <- lp_lin_iv(endog_data,
lags_endog_lin = 4,
shock = shock,
instrum = instrum,
use_twosls = TRUE,
trend = 0,
confint = 1.96,
hor = 20)
# Show all responses
plot(results_lin_iv)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.