Update on COVID-19 visuals for Nigeria

+++++++++++++++++++***************************+++++++++++++++++++++++++++

It is now 563 days since the first COVID-19 case was reported in Nigeria. As at September 13, 2021 the confirmed cases are 200,491 with 2,619 (1.31%) fatalities, however, 188,427 (94.43%) have recovered leaving 8,492 (4.26%) active cases.

Based on equal days forecast, by March 30, 2023, Nigeria’s aggregate confirmed COVID-19 cases are forecast to be:

Unconstrained forecast of COVID-19 for Nigeria

Model	Confirmed cases	Recoveries	Fatalities	Active	RMSE
With knots	1,827,374	1,725,589	23,939	77,846	157
Upper ARIMA	1,373,444	1,296,943	17,992	58,509	139
Without knots	446,360	421,498	5,847	19,015	373
Linear	285,695	269,782	3,743	12,171	367
Semilog	275,196	259,868	3,605	11,723	139
Essembled based on weight	230,722	217,871	3,022	9,829	388
Essembled with equal weight	219,645	207,411	2,877	9,357	272
Essembled based on weight of fit	175,520	165,744	2,299	7,477	1.6
Growth	3,490	3,296	46	149	162
Essembled based on summed weight	-80,806	-76,306	-1,059	-3,442	381
Quadratic Polynomial	-247,231	-233,460	-3,239	-10,532	200
Lower ARIMA	-879,781	-830,777	-11,525	-37,479	133
Smooth Spline	-1,489,237	-1,406,287	-19,509	-63,442	157

Constrained forecast of COVID-19 for Nigeria

Model	Confirmed cases	Recoveries	Fatalities	Active
ARIMA 95%	1,358,800	1,283,114	17,800.0	57,885
Without knots 95%	1,322,744	1,249,067	17,328.0	56,349
Essembled based on weight 95%	1,317,091	1,243,729	17,254.0	56,108
Essembled based on summed weight 95%	1,196,228	1,129,598	15,671.0	50,959
Smooth Spline 95%	949,604	896,711	12,440.0	40,453
Essembled with equal weight 95%	907,338	856,799	11,886.0	38,653
Essembled based on weight of fit 95%	691,295	652,790	9,056.0	29,449
Linear	285,695	269,782	3,743.0	12,171
Semilog	275,196	259,868	3,605.0	11,723
With knots 95%	209,277	197,620	2,742.0	8,915
Quadratic Polynomial 95%	93,768	88,545	1,228.0	3,995
Essembled with equal weight 80%	73,719	69,613	966.0	3,140
Quadratic Polynomial 80%	69,597	65,720	912.0	2,965
Essembled based on weight of fit 80%	69,156	65,304	906.0	2,946
Smooth Spline 80%	53,268	50,301	698.0	2,269
Essembled based on summed weight 80%	29,560	27,913	387.0	1,259
Essembled based on weight 80%	12,830	12,115	168.0	547
Without knots 80%	12,194	11,515	160.0	519
ARIMA 80%	6,389	6,034	84.0	272
Growth	3,490	3,296	46.0	149
With knots 80%	2	2	0.0	0

However, the actual forecast made by the various models on the last day i.e. March 30, 2023 is shown below:

Unconstrained forecasts on the last day

Constrained forecasts on the last day

Refer to Table 2 and Table 3 as well as Fig. 18-20 for more details on how the estimates and forecasts were obtained.

+++++++++++++++++++***************************+++++++++++++++++++++++++++

The visuals below supports this facts, take a look!

Fig. 1a Daily observed cases of COVID-19 in Nigeria

Fig. 2 Perecentages and previous days differences of COVID-19 in Nigeria Starting from February 29, 2020 to September 13, 2021

Fig. 3 Cases recorded in percentages Starting from February 29, 2020 to September 13, 2021 (legend as Fig. 2)

Fig. 4 Cumulative cases and Forecast of COVID-19 cases in Nigeria Starting from September 14, 2021 to March 30, 2023

Fig. 4a Components of COVID-19 cases in Nigeria between February 29, 2020 and September 13, 2021

Fig. 5 Number of days since average recorded cases exceeded one in each State

Fig. 6 Daily confirmed Coronavirus cases, by number of the days since 1 first daily case recorded

Fig. 7 Number of days from February 29, 2020 to September 13, 2021 that cases were recorded in the States

Fig. 7a Numbers of States that COVID-19 cases were reported on daily basis

Fig. 7b Forecast of Numbers of States that may still be COVID-19 affected by March 30, 2023

Coefficient estimates for number of states reporting COVID-19 daily

	Linear	Semilog	Growth	Spline without knots	Linear splines	Spline with knots	ARIMA
(Intercept)	15.510 ****	10.580 ****	2.566 ****	4.068 ****	2.807 *	0.097
	(0.535)	(1.493)	(0.055)	(0.899)	(1.582)	(2.283)
Series	-0.007 ****		-0.000 **
	(0.002)		(0.000)
log(Series)		0.576 **
		(0.277)
bs(Series, knots = NULL)1				34.920 ****
				(2.597)
bs(Series, knots = NULL)2				-6.030 ****
				(1.648)
bs(Series, knots = NULL)3				9.421 ****
				(1.419)
bs(DDPtable$Series, knots = c(30, 115, 247, 337, 420))1					-2.680
					(2.256)
bs(DDPtable$Series, knots = c(30, 115, 247, 337, 420))2					13.685 ****
					(1.673)
bs(DDPtable$Series, knots = c(30, 115, 247, 337, 420))3					22.948 ****
					(1.950)
bs(DDPtable$Series, knots = c(30, 115, 247, 337, 420))4					-1.000
					(1.729)
bs(DDPtable$Series, knots = c(30, 115, 247, 337, 420))5					29.435 ****
					(1.770)
bs(DDPtable$Series, knots = c(30, 115, 247, 337, 420))6					-12.723 ****
					(1.857)
bs(DDPtable$Series, knots = c(30, 115, 247, 337, 420))7					12.776 ****
					(1.960)
bs(DDPtable$Series, knots = c(30, 115, 247, 337, 420))8					13.417 ****
					(1.824)
bs(Series, knots = c(30, 247, 420))1						4.784 *
						(2.818)
bs(Series, knots = c(30, 247, 420))2						25.225 ****
						(2.420)
bs(Series, knots = c(30, 247, 420))3						9.860 ****
						(2.795)
bs(Series, knots = c(30, 247, 420))4						20.452 ****
						(2.524)
bs(Series, knots = c(30, 247, 420))5						-3.979
						(2.656)
bs(Series, knots = c(30, 247, 420))6						24.088 ****
						(2.606)
ma1							-0.753
							(0.028)
nobs	545	545	545	545	545	545	544
r.squared	0.029	0.008	0.010	0.307	0.758	0.453
adj.r.squared	0.027	0.006	0.008	0.303	0.754	0.446
sigma	6.239	6.307	0.637	5.282	3.136	4.707	3.202
statistic	16.326	4.333	5.376	79.817	209.737	74.134
p.value	0.000	0.038	0.021	0.000	0.000	0.000
df	1.000	1.000	1.000	3.000	8.000	6.000
logLik	-1770.119	-1776.026	-526.578	-1678.333	-1391.685	-1613.997	-1404.933
AIC	3546.239	3558.052	1059.156	3366.666	2803.370	3243.994	2813.866
BIC	3559.141	3570.954	1072.058	3388.170	2846.378	3278.400	2822.464
deviance	21136.571	21599.707	220.369	15092.144	5271.182	11918.340
df.residual	543.000	543.000	543.000	541.000	536.000	538.000
nobs.1	545.000	545.000	545.000	545.000	545.000	545.000	544.000
**** p < 0.001; *** p < 0.01; ** p < 0.05; * p < 0.1.

Properties of the estimated coefficients

	Linear	Semilog	Growth	Spline with knots	Linear splines	Spline without knots	Smooth spline	ARIMA
Absolute Error	2800	2900	6100	2100	1300	2400	1100	1300
Absolute Percent Error	440	420	420	230	140	280	120	130
Accuracy	0	0	0	0	0	0	0	0
Adjusted R Square	0.027	0.0061	0.008	0.45	0.75	0.3	0	0
Akaike’s Information Criterion AIC	3500	3600	1100	3200	2800	3400	0	2800
Allen’s Prediction Sum-Of-Squares (PRESS, P-Square)	0	0	0	0	0	0	0	0
Area under the ROC curve (AUC)	0.87	0.13	0.87	0.033	0.049	0.022	0.015	0.028
Average Precision at k	0	0	0	0	0	0	0	0
Bias	-0.00000000000000096	0.00000000000000074	11	-0.00000000000000038	0.000000000000000011	-0.0000000000000000078	0.0000000000000063	0.099
Brier score	40	40	200	20	10	30	8	10
Classification Error	1	1	1	1	1	1	1	1
F1 Score	0	0	0	0	0	0	0	0
fScore	0	0	0	0	0	0	0	0
GINI Coefficient	0.2	-0.2	0.2	0.7	0.9	0.6	0.9	0.9
kappa statistic	0	0	0	0	0	0	0	0
Log Loss	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf
Mallow’s cp	2	2	2	7	9	4	0	0
Matthews Correlation Coefficient	0	0	0	0	0	0	0	0
Mean Log Loss	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf
Mean Absolute Error	5.2	5.3	11	3.8	2.3	4.4	2.1	2.4
Mean Absolute Percent Error	0.8	0.78	0.77	0.41	0.26	0.52	0.21	0.24
Mean Average Precision at k	0	0	0	0	0	0	0.17	0
Mean Absolute Scaled Error	1.8	1.9	4	1.3	0.83	1.5	0.73	0.85
Median Absolute Error	4.9	5.1	12	3.3	1.9	4	1.6	2
Mean Squared Error	39	40	160	22	9.7	28	7.8	10
Mean Squared Log Error	0.32	0.31	2	0.15	0.075	0.19	0.052	0.069
Model turning point error	328	295	328	304	302	308	306	403
Negative Predictive Value	0	0	0	0	0	0	0	0
Percent Bias	-0.55	-0.52	0.72	-0.2	-0.1	-0.3	-0.081	-0.069
Positive Predictive Value	0	0	0	0	0	0	0	0
Precision	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
R Square	0.029	0.0079	0.0098	0.45	0.76	0.31	0	0
Relative Absolute Error	0.97	1	2.1	0.71	0.44	0.82	0.39	0.45
Recall	15	12	2.5	3	3.4	5.9	2.6	2.2
Root Mean Squared Error	6.2	6.3	13	4.7	3.1	5.3	2.8	3.2
Root Mean Squared Log Error	0.56	0.56	1.4	0.38	0.27	0.43	0.23	0.26
Root Relative Squared Error	0.99	1	2	0.74	0.49	0.83	0.44	0.51
Relative Squared Error	0.97	0.99	4.1	0.55	0.24	0.69	0.19	0.26
Schwarz’s Bayesian criterion BIC	3600	3600	1100	3300	2800	3400	0	2800
Sensitivity	0	0	0	0	0	0	0	0
specificity	0	0	0	0	0	0	0	0
Squared Error	21000	22000	90000	12000	5300	15000	4200	5600
Squared Log Error	170	170	1100	79	41	100	29	37
Symmetric Mean Absolute Percentage Error	0.42	0.44	1.3	0.32	0.22	0.37	0.19	0.21
Sum of Squared Errors	21000	22000	90000	12000	5300	15000	4200	5600
True negative rate	0	0	0	0	0	0	0	0
True positive rate	0	0	0	0	0	0	0	0

Fig. 8 Number of days the last COVID19 case was recorded as at September 13, 2021

Fig. 9 Diverging Bars of COVID-19 cases in the States (normalised)

Fig. 10 Monthly summary of COVID19 cases in the States

Fig. 11 Daily recorded cases of COVID19 in the States

Fig. 12 Number of days since average recorded cases exceeded one in each geo-political zone

Fig. 13 Number of days from February 29, 2020 to September 13, 2021 that cases were recorded in the Zones

Fig. 14 Monthly summary of COVID19 cases in the Zones

Table 1 Regression estimates of factors that might have stressed Nigeria’s economy within the period of the COVID-19 pandemic

	COVID19	PMS	Inflation	Teledensity	Birth rate	Population
(Intercept)	-300.967	-59.196	-225.229	375.635	-35.252 ****	15.876 ****
	(310.325)	(1845.262)	(201.385)	(428.828)	(7.996)	(0.323)
PMSprice	-0.123 ***		0.100 ****	0.196 ****	0.002	-0.000
	(0.033)		(0.015)	(0.036)	(0.002)	(0.001)
Inflation	0.998 ***	7.694 ****		-1.499 ***	-0.032 **	0.007
	(0.308)	(1.130)		(0.391)	(0.013)	(0.005)
Teledensity	0.604 ****	3.437 ****	-0.341 ***		-0.002	-0.002
	(0.108)	(0.641)	(0.089)		(0.007)	(0.003)
`Birth rate`	2.537	44.290	-9.390 **	-2.357		0.275 ****
	(6.900)	(38.116)	(3.791)	(9.505)		(0.055)
Population	14.578	-21.584	17.124	-16.940	2.320 ****
	(19.754)	(115.754)	(12.382)	(27.288)	(0.467)
COVID19		-4.059 ***	0.429 ***	1.141 ****	0.004	0.003
		(1.091)	(0.132)	(0.205)	(0.010)	(0.003)
nobs	20	20	20	20	20	20
r.squared	0.739	0.864	0.892	0.818	0.741	0.753
adj.r.squared	0.646	0.815	0.853	0.754	0.648	0.665
sigma	1.309	7.536	0.858	1.798	0.050	0.017
statistic	7.921	17.763	23.007	12.625	8.001	8.526
p.value	0.001	0.000	0.000	0.000	0.001	0.001
df	5.000	5.000	5.000	5.000	5.000	5.000
logLik	-30.199	-65.205	-21.756	-36.551	34.919	56.240
AIC	74.398	144.409	57.513	87.102	-55.839	-98.480
BIC	81.368	151.379	64.483	94.072	-48.868	-91.510
deviance	23.993	794.981	10.314	45.284	0.036	0.004
df.residual	14.000	14.000	14.000	14.000	14.000	14.000
nobs.1	20.000	20.000	20.000	20.000	20.000	20.000
**** p < 0.001; *** p < 0.01; ** p < 0.05; * p < 0.1.

Fig. 15 COVID-19, inflation, PMS price, Teledensity and birth rate in Nigeria

Fig. 15 Relationship between COVID-19, inflation, PMS price, Teledensity and birth rate in Nigeria

Table 2 The coefficient estimates of the various models for forecasting of COVID-19 for Nigeria¹

	Linear	Semilog	Growth	Spline without knots	Spline with knots	ARIMA	Quardratic polynomial	Essembled based on weight	Essembled based on summed weight
(Intercept)	280.308 ****	-118.185	4.547 ****	-130.898 **	-3.509		41.763	25.806	-1.883
	(32.821)	(89.353)	(0.132)	(61.732)	(62.444)		(47.504)	(148.945)	(19.207)
Day	0.269 ***		0.002 ****				2.802 ****
	(0.101)		(0.000)				(0.389)
log(Day)		88.811 ****
		(16.458)
bs(Day, knots = NULL)1				1228.198 ****
				(178.364)
bs(Day, knots = NULL)2				219.535 *
				(113.183)
bs(Day, knots = NULL)3				501.988 ****
				(97.478)
bs(Day, knots = BREAKS)1					5.730
					(120.718)
bs(Day, knots = BREAKS)2					12.469
					(78.199)
bs(Day, knots = BREAKS)3					844.874 ****
					(89.645)
bs(Day, knots = BREAKS)4					169.101 **
					(76.161)
bs(Day, knots = BREAKS)5					205.045 **
					(83.423)
bs(Day, knots = BREAKS)6					-86.771
					(78.502)
bs(Day, knots = BREAKS)7					2046.449 ****
					(78.165)
bs(Day, knots = BREAKS)8					-292.551 ****
					(78.824)
bs(Day, knots = BREAKS)9					85.904
					(91.572)
bs(Day, knots = BREAKS)10					392.347 ****
					(89.295)
bs(Day, knots = BREAKS)11					703.663 ****
					(81.561)
ar1						1.016
						(0.062)
ar2						-0.204
						(0.053)
ma1						-1.738
						(0.046)
ma2						0.815
						(0.036)
I(Day^2)							-0.004 ****
							(0.001)
fitf								0.828	-0.128
								(7.401)	(0.082)
fit0f								0.020	-0.010
								(1.037)	(0.090)
fit1f								-0.644	1.286 ****
								(6.216)	(0.127)
fitpif								-0.257	0.010
								(1.283)	(0.102)
fitaf								0.888	-0.152
								(5.577)	(0.093)
fitf:fit0f								-0.006
								(0.023)
fitf:fit1f								0.024
								(0.019)
fit0f:fit1f								0.015
								(0.025)
fitf:fitpif								-0.001
								(0.017)
fit0f:fitpif								0.001
								(0.002)
fit1f:fitpif								-0.011
								(0.020)
fitf:fitaf								-0.057
								(0.035)
fit0f:fitaf								-0.007
								(0.019)
fit1f:fitaf								0.031
								(0.032)
fitpif:fitaf								0.012
								(0.016)
fitf:fit0f:fit1f								-0.000
								(0.000)
fitf:fit0f:fitpif								0.000
								(0.000)
fitf:fit1f:fitpif								-0.000
								(0.000)
fit0f:fit1f:fitpif								-0.000
								(0.000)
fitf:fit0f:fitaf								0.000 *
								(0.000)
fitf:fit1f:fitaf								0.000
								(0.000)
fit0f:fit1f:fitaf								-0.000
								(0.000)
fitf:fitpif:fitaf								0.000
								(0.000)
fit0f:fitpif:fitaf								-0.000
								(0.000)
fit1f:fitpif:fitaf								-0.000
								(0.000)
fitf:fit0f:fit1f:fitpif								0.000
								(0.000)
fitf:fit0f:fit1f:fitaf								-0.000
								(0.000)
fitf:fit0f:fitpif:fitaf								-0.000 *
								(0.000)
fitf:fit1f:fitpif:fitaf								-0.000
								(0.000)
fit0f:fit1f:fitpif:fitaf								0.000
								(0.000)
fitf:fit0f:fit1f:fitpif:fitaf								-0.000
								(0.000)
nobs	563	563	563	563	563	562	563	563	563
r.squared	0.013	0.049	0.040	0.116	0.828		0.086	0.884	0.874
adj.r.squared	0.011	0.048	0.038	0.111	0.825		0.083	0.877	0.873
sigma	388.863	381.541	1.560	368.632	163.682	157.544	374.385	137.042	139.532
statistic	7.106	29.119	23.384	24.391	241.403		26.448	130.425	771.076
p.value	0.008	0.000	0.000	0.000	0.000		0.000	0.000	0.000
df	1.000	1.000	1.000	3.000	11.000		2.000	31.000	5.000
logLik	-4155.157	-4144.455	-1048.113	-4124.072	-3662.928	-3639.618	-4133.293	-3552.513	-3576.107
AIC	8316.314	8294.911	2102.226	8258.144	7351.856	7289.236	8274.586	7171.026	7166.213
BIC	8329.314	8307.911	2115.225	8279.811	7408.189	7310.893	8291.919	7314.024	7196.546
deviance	84831112.306	81666701.075	1364.736	75962321.679	14762196.420		78491708.762	9972507.649	10844362.905
df.residual	561.000	561.000	561.000	559.000	551.000		560.000	531.000	557.000
nobs.1	563.000	563.000	563.000	563.000	563.000	562.000	563.000	563.000	563.000
**** p < 0.001; *** p < 0.01; ** p < 0.05; * p < 0.1.

Table 3 Various selection criteria for the estimated models²

	Linears	Semilogs	Growths	Spline with knots	Spline without knots	Smooth spline	ARIMA	Quardratic polynomial	Essembled equal weight	Essembled on weight	Essembled summed weight	Essembled weight of fit
Absolute Error	160000	160000	200000	60000	150000	44000	51000	160000	200000	44000	44000	200000
Absolute Percent Error	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf
Accuracy	0	0	0	0	0	0	0	0	0.032	0	0	0.032
Adjusted R Square	0.011	0.048	0.038	0.82	0.11	0	0	0.083	0	0.88	0.87	0
Akaike’s Information Criterion AIC	8300	8300	2100	7400	8300	0	7300	8300	0	7200	7200	0
Allen’s Prediction Sum-Of-Squares (PRESS, P-Square)	0	0	0	0	0	0	0	0	0	0	0	0
Area under the ROC curve (AUC)	0.008	0.008	0.008	0.085	0.008	0.008	0.016	0.008	0.008	0.032	0.008	0.008
Average Precision at k	0	0	0	0	0	0	0	0	0.5	0	0	0.5
Bias	0.00000000000000081	0.00000000000001	350	-0.0000000000000022	-0.000000000000034	-0.00000000000043	1.9	0.000000000000032	360	-0.0000000000000019	0.000000000000001	360
Brier score	200000	100000	300000	30000	100000	20000	20000	100000	NaN	20000	20000	NaN
Classification Error	1	1	1	1	1	1	1	1	1	1	1	1
F1 Score	0	0	0	0	0	0	0	0	0	0	0	0
fScore	0	0	0	0	0	0	0	0	0	0	0	0
GINI Coefficient	0.1	0.1	0.1	0.9	0.3	1	0.9	0.5	0.9	1	1	0.9
kappa statistic	0	0	0	0	0	0	0	0	NaN	0	0	NaN
Log Loss	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf
Mallow’s cp	2	2	2	12	4	0	0	3	0	0	6	0
Matthews Correlation Coefficient	0	0	0	0	0	0	0	0	NaN	0	0	NaN
Mean Log Loss	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	NaN	Inf	Inf	NaN
Mean Absolute Error	290	280	350	110	270	78	90	290	NaN	78	79	NaN
Mean Absolute Percent Error	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf	Inf
Mean Average Precision at k	0	0	0	0	0	0.5	0	0	0	0	0	0
Mean Absolute Scaled Error	2.8	2.7	3.4	1	2.6	0.75	0.86	2.7	3.4	0.74	0.75	3.4
Median Absolute Error	250	220	210	73	250	35	43	270	110	41	37	150
Mean Squared Error	NA	NA	NA	NA	NA	NA	NA	NA	NaN	NA	NA	NaN
Mean Squared Log Error	3	NaN	13	NaN	NaN	NaN	0.26	2.4	29	NaN	NaN	29
Model turning point error	286	286	286	278	288	271	374	278	344	269	235	341
Negative Predictive Value	0	0	0	0	0	0	0	0	0	0	0	0
Percent Bias	-Inf	NaN	-Inf	NaN	NaN	NaN	-Inf	-Inf	NaN	NaN	NaN	NaN
Positive Predictive Value	0	0	0	0	0	0	0	0	0	0	0	0
Precision	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	NaN	NaN	0
R Square	0.013	0.049	0.04	0.83	0.12	0	0	0.086	0	0.88	0.87	0
Relative Absolute Error	0.99	0.96	1.2	0.36	0.92	0.27	0.31	0.98	0.47	0.27	0.27	0.63
Recall	280	-12	4.6	-2.1	-99	-0.2	0.5	58	0	3.8	-0.38	0
Root Mean Squared Error	NA	NA	NA	NA	NA	NA	NA	NA	NaN	NA	NA	NaN
Root Mean Squared Log Error	1.7	NaN	3.7	NaN	NaN	NaN	0.51	1.5	5.3	NaN	NaN	5.3
Root Relative Squared Error	0.99	0.98	1.3	0.41	0.94	0.36	0.4	0.96	1.4	0.34	0.36	1.4
Relative Squared Error	0.99	0.95	1.8	0.17	0.88	0.13	0.16	0.91	1.8	0.12	0.13	1.8
Schwarz’s Bayesian criterion BIC	8300	8300	2100	7400	8300	0	7300	8300	0	7300	7200	0
Sensitivity	0	0	0	0	0	0	0	0	0.5	0	0	0.5
specificity	0	0	0	0	0	0	0	0	0	0	0	0
Squared Error	85000000	82000000	160000000	15000000	76000000	11000000	14000000	78000000	160000000	10000000	11000000	160000000
Squared Log Error	1700	NaN	7500	NaN	NaN	NaN	150	1300	16000	NaN	NaN	16000
Symmetric Mean Absolute Percentage Error	0.89	0.89	1.8	0.58	0.85	0.34	0.4	0.92	0.6	0.38	0.36	0.7
Sum of Squared Errors	85000000	82000000	160000000	15000000	76000000	11000000	14000000	78000000	160000000	10000000	11000000	160000000
True negative rate	0	0	0	0	0	0	0	0	0	0	0	0
True positive rate	0	0	0	0	0	0	0	0	0	0	0	0

Fig. 18 Models of COVID-19 Cases using ensemble technology

Fig. 19 Equal duration forecast of COVID-19 Cases from the ensemble models using a native plotting method

Fig. 20 Unconstrained forecast of COVID-19 cases in Nigeria Starting from September 14, 2021 to March 30, 2023 using a more advaced plotting method

Fig. 20a

Fig. 20b Constrained forecast of COVID-19 cases in Nigeria Starting from September 14, 2021 to March 30, 2023 using a more advaced plotting method

Fig. 20b Lagged forecasts (1-14 days) of COVID-19 cases in Nigeria

Table 4 Summary of the dynamic COVID-19 forecasts

Model	Sep 01, 21 - Mar 03, 23	Sep 08, 21 - Mar 17, 23	Sep 11, 21 - Mar 23, 23	Sep 12, 21 - Mar 25, 23	Sep 13, 21 - Mar 27, 23	Sep 14, 21 - Mar 29, 23
Without knots	146852	320476	399964	413515	433712	439196
Smooth Spline	4149440	3054018	2781952	2544669	2400599	2094742
With knots	-1025406	-863720	53801	-373943	-356480	-1040242
Quadratic Polynomial	-333492	-283390	-259126	-255514	-249484	-248652
Lower ARIMA	-821057	-822030	-764086	-798187	-795611	-848206
Upper ARIMA	1364751	1395234	1454278	1428825	1435210	1397877
Essembled with equal weight	246625	290024	304407	326777	299916	275403
Essembled based on weight	126662	369475	430732	544769	482432	505578
Essembled based on summed weight	-23958	-60163	-60515	-80576	-77561	-65417
Essembled based on weight of fit	170950	207234	219757	234242	219948	207023

Table 3 RMSE of the models in the dynamic forecasts

Model	Sep 01, 21 - Mar 03, 23	Sep 08, 21 - Mar 17, 23	Sep 11, 21 - Mar 23, 23	Sep 12, 21 - Mar 25, 23	Sep 13, 21 - Mar 27, 23	Sep 14, 21 - Mar 29, 23
Without knots	369.27	368.88	368.51	368.21	367.97	367.65
Smooth Spline	157.48	160.22	160.31	160.72	160.82	161.47
With knots	138.9	139.76	139.47	139.5	139.42	139.53
Quadratic Polynomial	372.33	373.58	374.06	373.91	373.89	373.63
Lower ARIMA	156.56	157.2	156.99	156.95	156.81	156.89
Upper ARIMA	156.56	157.2	156.99	156.95	156.81	156.89
Essembled with equal weight	199.57	200.34	200.23	200.19	200.1	200.13
Essembled based on weight	413.06	412.12	411.03	410.59	410.2	409.9
Essembled based on summed weight	388.65	386.66	385.8	385.44	385.11	384.75
Essembled based on weight of fit	205.78	201.75	199.82	199.58	199.13	199.17

Fig. 21 Percentage of COVID19 cases that resulted into casualty per State as at September 13, 2021

Fig. 22 Percentage of COVID19 cases that recovered per State as at September 13, 2021

Fig. 23 Percentage of recoveries and deaths from COVID19 cases per Zone as at September 13, 2021

Fig. 24 Distribution of COVID19 in the States as at September 13, 2021

---

1. Note that Smooth spline does not estimate coefficients and other model-characteristic statistics↩︎

2. Multiclass Area Under the Curve (MAUC) is the mean of the various AUC estimated from the same data↩︎