Estimate regime-dependent local projections

Source: R/lp.R

Estimate a regime-dependent local projection (i.e. a state-dependent LP), with an exogenous state indicator, of the specification: $$Y_{t+h} = X_t \beta_{s_t} + \epsilon_t$$ where t is the time index, and s is a mutually exclusive state of the world observed at time t. When the regime vector is not supplied by the user, then a two-state regime series is estimated via random forest.

RLP(
  data,
  horizons = 1,
  freq = "month",
  type = "const",
  p = 1,
  lag.ic = NULL,
  lag.max = NULL,
  NW = FALSE,
  NW_lags = NULL,
  NW_prewhite = NULL,
  regime = NULL,
  regime.method = "rf",
  regime.n = 2
)

Arguments

data

data.frame, matrix, ts, xts, zoo: Endogenous regressors
horizons

int: forecast horizons
freq

string: frequency of data ('day', 'week', 'month', 'quarter', or 'year')
type

string: type of deterministic terms to add ('none', 'const', 'trend', or 'both')
p

int: lags
lag.ic

string: information criterion to choose the optimal number of lags ('AIC' or 'BIC')
lag.max

int: maximum number of lags to test in lag selection
NW

boolean: Newey-West correction on variance-covariance matrix
NW_lags

int: number of lags to use in Newey-West correction
NW_prewhite

boolean: TRUE prewhite option for Newey-West correction (see sandwich::NeweyWest)
regime

string: name or regime assignment vector in the design matrix (data)
regime.method

string: regime assignment technique ('rf', 'kmeans', 'EM', 'BP')
regime.n

int: number of regimes to estimate (applies to kmeans and EM)

Value

list of lists, one list per regime, each regime with objects with elements data, model, forecasts, residuals; if there is more than one forecast horizon estimated, then model, forecasts, residuals will each be a list where each element corresponds to a single horizon

References

  1. Jorda, Oscar "Estimation and Inference of Impulse Responses by Local Projections" 2005.

See also

LP()

lp_irf()

RLP()

rlp_irf()

Examples

# \donttest{

  # simple time series
  AA = c(1:100) + rnorm(100)
  BB = c(1:100) + rnorm(100)
  CC = AA + BB + rnorm(100)
  date = seq.Date(from = as.Date('2000-01-01'), by = 'month', length.out = 100)
  Data = data.frame(date = date, AA, BB, CC)
  # add regime
  Data = dplyr::mutate(Data, reg = dplyr::if_else(AA > median(AA), 1, 0))

  # local projection forecasts
  rlp =
    sovereign::RLP(
      data = Data,
      regime = 'reg',
      horizon = c(1:10),
      freq = 'month',
      p = 1,
      type =  'const',
      NW = TRUE,
      NW_lags = 1,
      NW_prewhite = FALSE)

 # impulse response function
 rirf = sovereign::rlp_irf(rlp)

# }