Skip to main content
. 2022 Mar 1;53(9):3920–3931. doi: 10.1017/S0033291722000587

Table 3.

Parameter estimates of the selected ARIMA models for weekly COVID-mention among help-seekers in Open Up (Open Up, CMD, and No CMD)

Weekly ARIMA model Parameter Estimate Standard error p AIC SBC
Open Upa AR(1,0,0) MUb −6.34 12.77 0.620 526 532
AR1c 0.77 0.09 <0.001
COVIDd 0.08 0.03 0.002
AR(1,0,0) MU −4.41 9.79 0.652 520 528
AR1 0.71 0.10 <0.001
ImpF-2e 34.87 17.14 0.042
MagI-3f 0.14 0.03 <0.001
CMDg AR(1,0,0) MU −4.68 10.03 0.641 497 503
AR1 0.77 0.09 <0.001
COVID 0.06 0.02 0.003
AR(1,0,0) MU −3.20 7.67 0.676 492 500
AR1 0.71 0.09 <0.001
ImpF-2 26.53 13.41 0.048
MagI-3 0.10 0.03 <0.001
No CMDh AR(1,0,0) MU −0.78 2.35 0.740 395 401
AR1 0.59 0.11 <0.001
COVID 0.02 0.01 0.004
AR(1,0,0) MU −0.61 2.04 0.763 391 397
AR1 0.54 0.12 <0.001
MagI-3 0.04 0.01 <0.001
a

Selected ARIMA model for COVID-mention.

b

Intercept.

c

Autoregressive, lagged by one day.

d

COVID-19 reported cases in Hong Kong.

e

Impulse function for the 2nd wave: the value of 1 for the 2nd wave of COVID-19.

f

Magnitude effect for the 3rd wave: COVID-19 magnitude in the 3rd wave of outbreak.

g

Selected ARIMA model for the CMD-mention subgroup.

h

Selected ARIMA model for non-CMD mention subgroup.