Non linear autoregressive model
LSTAR model
Coefficients:
Low regime:
const.L phiL.1 phiL.2 phiL.3 phiL.4
35.0897740 0.3357427 1.0552361 0.5757245 -1.7592532
High regime:
const.H phiH.1 phiH.2 phiH.3 phiH.4
-23.0191951 0.3969194 -0.3671385 -0.4521622 1.3455011
Smoothing parameter: gamma = 3.398
Threshold
Variable: external
Value: 163.6
Residuals:
Min 1Q Median 3Q Max
-27.6013 -9.6260 -1.3961 9.5330 45.6929
Fit:
residuals variance = 176, AIC = 732, MAPE = 12.07%
Coefficient(s):
Estimate Std. Error t value Pr(>|z|)
const.L 35.08977 3.62600 9.6773 < 2.2e-16 ***
phiL.1 0.33574 0.08201 4.0939 4.242e-05 ***
phiL.2 1.05524 0.11607 9.0916 < 2.2e-16 ***
phiL.3 0.57573 0.12741 4.5187 6.221e-06 ***
phiL.4 -1.75925 0.19593 -8.9791 < 2.2e-16 ***
const.H -23.01919 10.21611 -2.2532 0.024245 *
phiH.1 0.39692 0.16672 2.3808 0.017275 *
phiH.2 -0.36714 0.22588 -1.6254 0.104087
phiH.3 -0.45216 0.25449 -1.7767 0.075616 .
phiH.4 1.34550 0.22836 5.8921 3.814e-09 ***
gamma 3.39803 1274.10784 0.0027 0.997872
th 163.57269 54.89853 2.9795 0.002887 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Non-linearity test of full-order LSTAR model against full-order AR model
F = 25.728 ; p-value = 7.8378e-15
Threshold
Variable: external>


雷达卡


京公网安备 11010802022788号







