Report 3 Bangladesh

This report is created based on a publicly available data. The data are available on a Google Sheet. Please see the Data Source section for link. Please note that publicly available data are not official and MAY BE UNRELIABLE. I personally did not verify them. I am using them for educational purposes. Use at your discretion.

3.1 Epidemic curve and search for peak

The epidemic curve for Bangladesh and a tentative peak. The red line in the plot indicates a potential peak based on available data. IF the peak (red line) is at the far right, then we may not have reached a peak yet. We must have sustained decrease of incidences to be sure about any peak.

We fit a log linear model to estimate the doubling time (growh phase) and halving time (decline phase, when that happens). The estimated fitted line (solid line) with 95% confidence interval (dotten lines) are superimposed on the epidemic curve below.

The current doubling time as of 2020-09-10 is 9.9 days with a 95% confidence interval of (9.1, 10.9) days.

3.1.1 Estimating the Overall Reproduction Number, $R_{0}$

The log-linear model estimates the $R_{0}$ as follows:

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   1.511   1.567   1.591   1.592   1.612   1.679

3.1.2 Estimating the Effectuve Reproduction Number ( $R_{e}$ )

The estimation of effective reproduction number involves the distribution of serial interval (SI). Based on the literature (REF to be added), a gamma distribution with mean 7.5 and standard deviation 3.5

3.1.3 Simulating the $R_{e}$ on Weekly Sliding Window

Now we perform a simulation of different SI distribution to obtain the $R_{e}$ . For this, we vary the parameters of gamma distribution and obtain the following.

# Recalculating with varying SI distribution parameters 
# we get these results:

config = make_config(list(mean_si = 7.5, std_mean_si = 3.5, 
                                       min_mean_si = 2, max_mean_si =10.0,
                                       std_si = 3.5, std_std_si = 1,
                                       min_std_si = 0.5, max_std_si = 4))
bd_res_uncertain_si <- 
  estimate_R(bd_confirmed_cases$daily_cases, method = "uncertain_si", 
             config = config)

plot_Ri(bd_res_uncertain_si)

Table 3.1: Effective Reproduction Number by Week as of 2020-07-03
	t_start	t_end	Mean(R)	Std(R)	Quantile.0.025(R)	Quantile.0.975(R)
105	106	112	1.046632	0.0169938	1.0189608	1.081899
106	107	113	1.049673	0.0145129	1.0231781	1.079145
107	108	114	1.064205	0.0138124	1.0355469	1.090588
108	109	115	1.067638	0.0154839	1.0303541	1.093207
109	110	116	1.072344	0.0178503	1.0300534	1.099355
110	111	117	1.066645	0.0202760	1.0205370	1.096095
111	112	118	1.027756	0.0209605	0.9843239	1.059694

The effective reproduction number based on the last week’s incidences show an average of 1.03 with a 95% confidence interval (0.98, 1.06).

3.1.4 Simulating $R_{e}$ on a Monthly Sliding Window

Table 3.2: Effective Reproduction Number by Month as of 2020-07-03
	t_start	t_end	Mean(R)	Std(R)	Quantile.0.025(R)	Quantile.0.975(R)
59	60	89	1.328111	0.1218987	1.093749	1.527731
60	61	90	1.326367	0.1225633	1.096821	1.529301
61	62	91	1.319118	0.1219674	1.093018	1.523864
62	63	92	1.312845	0.1204491	1.092004	1.517296
63	64	93	1.306606	0.1180362	1.090871	1.508910
64	65	94	1.303426	0.1154575	1.092073	1.502786
65	66	95	1.296084	0.1126439	1.088118	1.490242
66	67	96	1.289401	0.1097843	1.086316	1.478005
67	68	97	1.284089	0.1072704	1.085707	1.467915
68	69	98	1.268620	0.1041213	1.075342	1.446044
69	70	99	1.256148	0.1003377	1.072203	1.427868
70	71	100	1.248356	0.0963948	1.073534	1.414545
71	72	101	1.247968	0.0931521	1.078450	1.409307
72	73	102	1.243734	0.0904911	1.077346	1.399898
73	74	103	1.241699	0.0882930	1.077199	1.392902
74	75	104	1.224717	0.0856263	1.063337	1.370345
75	76	105	1.206741	0.0819941	1.054814	1.345852
76	77	106	1.195827	0.0779376	1.055362	1.329129
77	78	107	1.182595	0.0735190	1.051625	1.310244
78	79	108	1.174473	0.0690322	1.052361	1.295629
79	80	109	1.162138	0.0643547	1.047242	1.275574
80	81	110	1.168266	0.0606673	1.060065	1.275402
81	82	111	1.168308	0.0584652	1.058817	1.269725
82	83	112	1.157930	0.0568349	1.048442	1.254106
83	84	113	1.146390	0.0549897	1.041503	1.238319
84	85	114	1.147418	0.0532516	1.048453	1.236274
85	86	115	1.135368	0.0515691	1.038929	1.221583
86	87	116	1.127179	0.0492675	1.037022	1.210468
87	88	117	1.116696	0.0464521	1.032996	1.196648
88	89	118	1.099904	0.0429453	1.023926	1.175051

The effective reproduction number based on the last months’s incidences show an average of 1.1 with a 95% confidence interval (1.02, 1.18).

3.1.5 Cumulative Deaths (official and unofficial)

Table 3.3: Total number of deaths as of 2020-09-09
country	Total Deaths
Bangladesh	4593
Bangladesh(unoff)	1217

3.2 Bangladesh in South Asia

3.2.1 Infection

Table 3.4: Total number of cases as of 2020-09-09
country	Total Cases
India	4465863
Bangladesh	331078
Pakistan	300030
Indonesia	203342
Bangladesh(unoff)	156391
Singapore	57166
Nepal	49219
Malaysia	9583
Sri Lanka	3147
Bhutan	234

3.2.2 Deaths

Table 3.5: Total number of deaths as of 2020-09-09
country	Total Deaths
India	75062
Indonesia	8336
Pakistan	6365
Bangladesh	4593
Bangladesh(unoff)	1217
Nepal	312
Malaysia	128
Singapore	27
Sri Lanka	12
Bhutan	0

3.2.3 Cases and deaths compared on a specific day

Bangladesh has entered into day 186 since first confirmed case.

Table 3.6: Bangladesh and its peers: total cases and deaths compared on day 186
Country	Total cases	Total deaths
India	1803695	38135
Indonesia	184268	7750
Pakistan	295053	6283
Bangladesh	331078	4593
Malaysia	8943	124
Nepal	19063	49
Singapore	50369	27
Sri Lanka	2814	11
Bhutan	233	0