lppl_estimate_rob_1s: 1st step LPPL estimation procedure by Geraskin and Fantazzini...
In deanfantazzini/bubble: Tests for financial bubbles and explosive behavior

Description Usage Arguments Details Value

This function performs the 1st step of the LPPL estimation procedure by Geraskin and Fantazzini (2013) and Fantazzini (2016)

lppl_estimate_rob_1s(
  x,
  par_start = c(bet = 0.5, ome = 6, A = log(x[NROW(x)]), B = 0, C1 = 0, C2 = 0),
  max.win.tc = 0.1
)

`x`	is a T x 1 data vector
`par_start`	is a 6 x 1 vector of starting values for the parameters to be estimated
`max.win.tc`	is a scalar setting the max window size (in percentage terms) used to fix the critical time tc in this estimation step

This function performs the 1st step of the LPPL estimation procedure by Geraskin and Fantazzini (2013) and Fantazzini (2016) using the LPPL formula by Filimonov and Sornette (2013): The critical time tc is fixed at tc = t2 + 0.1 x (t2-t1), where t2 and t1 are the last and the first observation of the estimation sample, respectively. The remaining LPPL parameters [A, B, C1, C2, beta, omega] are estimated by using a quasi-Newton method algorithm. The starting values for the parameter vector is given by:

par_start[1] = beta

par_start[2] = omega

par_start[3] = A

par_start[4] = B

par_start[5] = C1

par_start[6] = C2

par_est1 is a 6 x 1 vector of estimated parameters

deanfantazzini/bubble documentation built on Oct. 22, 2020, 2:43 p.m.