Cara mengimplementasikan model GARCH di Python

Model GARCH di Python

Chelsea Yang

Data Science Instructor

Paket Python "arch"

from arch import arch_model

¹ Kevin Sheppard. (2019, March 28). bashtage/arch: Release 4.8.1 (Version 4.8.1). Zenodo. http://doi.org/10.5281/zenodo.2613877

Alur kerja

Bangun model GARCH dalam tiga langkah:

Spesifikasikan model
Fit model
Buat ramalan

Spesifikasi model

Asumsi model:

Distribusi: "normal" (default), "t", "skewt"
Model mean: "constant" (default), "zero", "AR"
Model volatilitas: "GARCH" (default), "ARCH", "EGARCH"

basic_gm = arch_model(sp_data['Return'], p = 1, q = 1, 
                      mean = 'constant', vol = 'GARCH', dist = 'normal')

Fitting model

Tampilkan output fitting setiap n iterasi:

gm_result = gm_model.fit(update_freq = 4)

Proses fitting model

Matikan tampilan:

gm_result = gm_model.fit(disp = 'off')

Hasil fit: parameter

Diestimasi dengan "maximum likelihood method"

print(gm_result.params)

mu          0.077239
omega       0.039587
alpha[1]    0.167963
beta[1]     0.786467
Name: params, dtype: float64

Hasil fit: ringkasan

print(gm_result.summary())

Ringkasan hasil pemodelan

Hasil fit: plot

gm_result.plot()

Plot estimasi model

Peramalan model

# Peramalan 5 periode ke depan
gm_forecast = gm_result.forecast(horizon = 5)

# Cetak baris terakhir ramalan varians
print(gm_forecast.variance[-1:])

                 h.1       h.2       h.3       h.4       h.5
Date                                                        
2019-10-10  0.994079  0.988366  0.982913  0.977708  0.972741

h.1 pada baris "2019-10-10": ramalan 1-langkah ke depan menggunakan data hingga dan termasuk tanggal tersebut

Ayo berlatih!

Model GARCH di Python