key: cord-0287335-u7esax93
authors: Amiranashvili, A. G.; Khazaradze, K. R.; Japaridze, N. D.
title: The statistical analysis of daily data associated with different parameters of the New Coronavirus COVID-19 pandemic in Georgia and their two-week interval prediction in summer 2021
date: 2021-09-12
journal: nan
DOI: 10.1101/2021.09.08.21263265
sha: 43585b98ddef51e0c176f65fb478118f97b2d7e5
doc_id: 287335
cord_uid: u7esax93

The lockdown introduced in Georgia on November 28, 2020 brought positive results. There were clearly positive tendencies in the spread of COVID-19 to February - first half of March 2021. However, in April-May 2021 there was a significant deterioration in the epidemiological situation. From June to August 2021, the epidemiological situation with Covid-19 in Georgia became very difficult. In this work results of the next statistical analysis of the daily data associated with New Coronavirus COVID-19 infection of confirmed (C), recovered (R), deaths (D) and infection rate (I) cases of the population of Georgia in the period from June 01, 2021 to August 31, 2021 are presented. It also presents the results of the analysis of two-week forecasting of the values of C, D and I. As earlier, the information was regularly sent to the National Center for Disease Control & Public Health of Georgia and posted on the Facebook page https://www.facebook.com/Avtandil1948/. The analysis of data is carried out with the use of the standard statistical analysis methods of random events and methods of mathematical statistics for the non-accidental time-series of observations. In particular, the following results were obtained. Georgia's ranking in the world for Covid-19 monthly mean values of infection and deaths cases in summer 2021 (per 1 million population) was determined. Among 159 countries with population [≥] 1 million inhabitants in August 2021 Georgia was in the 1 place on new infection cases and on Death. A comparison between the daily mortality from Covid-19 in Georgia in summer 2021 with the average daily mortality rate in 2015-2019 shows, that the largest share value of D from mean death in 2015-2019 was 66.0 % (26.08.2021 and 31.08.2021), the smallest 6.0 % (09.07.2021). The statistical analysis of the daily and decade data associated with coronavirus COVID-19 pandemic of confirmed, recovered, deaths cases and infection rate of the population of Georgia are carried out. Maximum daily values of investigation parameters are following: C = 6208 (17.08.2021), R = 6177 (29.08.2021), D = 79 (26.08.2021 and 31.08.2021), I = 13.0 % (17.08.2021). Maximum mean decade values of investigation parameters are following: C = 5019 (2 Decade of August 2021), R = 4822 (3 Decade of August 2021), D = 69 (3 Decade of August 2021), I = 10.88 % (2 Decade of August 2021). It was found that as with September 2020 to February 2021 and in spring 2021 [7,8], from June to August 2021 the regression equations for the time variability of the daily values of C, R and D have the form of a tenth order polynomial. Mean values of speed of change of confirmed -V(C), recovered - V(R), deaths - V(D) and infection rate V(I) coronavirus-related cases in different decades of months in the summer 2021 were determined. Maximum mean decade values of investigation parameters are following: V(C) = +134 cases/day (1 Decade of August 2021), V(R) = +134 cases/day (2 Decade of August 2021), V(D) = +2.4 cases/day (3 Decade of August 2021), V(I) = + 0.25 %/ day (1 decades of August 2021). Cross-correlations analysis between confirmed COVID-19 cases with recovered and deaths cases shows, that the maximum effect of recovery is observed 19 days after infection (RC=0.95), and deaths - after 16 and 18 days (RC=0.94). In Georgia in the summer 2021, the duration of the impact of the delta variant of the coronavirus on people (recovery, mortality) could be up to two months. Comparison of real and calculated predictions data of C, D and I in Georgia are carried out. It was found that in summer 2021 two-week daily and mean two-week real values of C, D and I practically fall into the 67% - 99.99% confidence interval of these predicted values. With September 1, 2021, it is started monthly forecasting of C, D and I values. As earlier, the comparison of data about C and D in Georgia (GEO) with similar data in Armenia (ARM), Azerbaijan (AZE), Russia (RUS), Turkey (TUR) and in the World (WRL) is also carried out.

Twenty months has passed since the outbreak of the novel coronavirus in China, recognized on March 11, 2020 as a pandemic due to its rapid spread in the world [1] . Over the indicated period of time, despite the measures taken (including vaccination), the overall level of morbidity and mortality in many countries of the world remains quite high. Together with epidemiologists, scientists and specialists of various disciplines from all over the world continue intensive research of this unprecedented phenomenon (including Georgia [2] [3] [4] [5] [6] [7] [8] ).

So in our works [5] [6] [7] [8] it was noted that specialists in the field of physical and mathematical sciences make an important contribution to research on the spread of the new coronavirus COVID-19. Work on systematization, statistical analysis, forecasting, spatial-temporal modeling of the spread of the new coronavirus is being actively continued at the present time [9-19, etc.] .

This work is a continuation of the researches [5] [6] [7] [8] . In this work results of a statistical analysis of the daily data associated with New Coronavirus COVID-19 infection of confirmed (C), recovered (R), deaths (D) and infection rate (I) cases of the population of Georgia in the period from June 01, 2021 to August, 2021 are presented. It also presents the results of the analysis of two-week forecasting of the values of C, D and I. The information was regularly sent to the National Center for Disease Control & Public Health of Georgia and posted on the Facebook page https://www.facebook.com/Avtandil1948/.

The comparison of data about C and D in Georgia with similar data in Armenia, Azerbaijan, Russia, Turkey and in the world is also carried out.

We used standard methods of statistical analysis of random events and methods of mathematical statistics for non-random time series of observations [6-8, 20,21] .

The study area: Georgia. Data of John Hopkins COVID-19 Time Series Historical Data (with US State and County data) [https://www.soothsawyer.com/john-hopkins-time-series-data-with-us-state-and-countycity-detail-historical/; https://data.humdata.org/dataset/total-covid-19-tests-performed-by-country] and https://stopcov.ge about daily values of confirmed, recovered, deaths and infection rate coronavirus-related cases, from June 01, 2021 to August 31, 2021 are used. The work also used data of National Statistics Office of Georgia (Geostat) on the average monthly total mortality in Georgia in October-May 2015-2019 [https://www.geostat.ge/en/]. In the proposed work the analysis of data is carried out with the use of the standard statistical analysis methods of random events and methods of mathematical statistics for the nonaccidental time-series of observations [6-8, 20, 21] .

The following designations will be used below: Meanaverage values; Minminimal values; Max -maximal values; Range -Max-Min; St Dev -standard deviation; σm -standard error; CV = 100·St Dev/Mean coefficient of variation, %; R 2coefficient of determination; rcoefficient of linear correlation; CRcoefficient of cross correlation; Lag = 1, 2…30 Day; KDW -Durbin-Watson statistic; Calccalculated data; Real -measured data; D1, D2, D3numbers of the month decades; α -the level of significance; C, R, Ddaily values of confirmed, recovered and deaths coronavirus-related cases; V(C), V(R) and V(D) -daily values of speed of change of confirmed, recovered and deaths coronavirus-related cases (cases/day); I -daily values of infection rate (or positive rate) coronavirus-related cases (100· C/number of coronavirus tests performed); V(I) -daily values of speed of change of I; DCdeaths coefficient, % = (100· D/C). Official data on number of coronavirus tests performed are published from December 05, 2020 [https://stopcov.ge].

The statistical programs Data Fit 7, Mesosaur and Excel 16 were used for calculations. The curve of trend is equation of the regression of the connection of the investigated parameter with the time at the significant value of the determination coefficient and such values of KDW, where the residual values are accidental. If the residual values are not accidental the connection of the investigated parameter with the time we will consider as simply regression.

The calculation of the interval prognostic values of C, D and I taking into account the periodicity in the time-series of observations was carried out using Excel 16 (the calculate methodology was description in [6] In the Table 1 [6] the scale of comparing real data with the predicted ones and assessing the stability of the time series of observations in the forecast period in relation to the pre-predicted one (period for prediction calculating) is presented.

The results in the Fig. 1-24 

The time-series curves and statistical characteristics of Covid-19 confirmed and deaths cases (to 1 million populations) in Georgia, its neighboring countries and World from June 01, 2021 to August 31, 2021 in Fig. 1 -2 and in Table 1 -2 are presented. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint Significant linear correlation (rmin = ± 0.21, α = 0.05) between these countries on C value varies from 0.21 (pair Armenia-Russia) to 0.92 (pair Armenia-Azerbaijan). Linear correlation between World and these countries is significant and varied from 0.31 (pair World-Russia) to 0.76 (pair World-Georgia). Thus, in summer 2021 in these countries there were more or less similar trends to the world the development of the coronavirus epidemic ( Table 1) .

Variability of the values of D per 1 million populations is as follows (Fig. 2 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Significant linear correlation between these countries on D value varies from 0.27 (pair Russia-Turkey) to 0.92 (pair Georgia-Azerbaijan and pair Georgia -Turkey). Linear correlation between World and these countries besides Armenia and Russia is significant and varied from 0.21 (pair World-Azerbaijan) to 0.33 (pair World-Turkey). Thus, for Georgia, Azerbaijan and Turkey in summer 2021 there were some similar trends with world in the variability of mortality from coronavirus (Table 2 ).

In Table 3 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint As follows from this Table range of change of mean monthly values of C for 159 countries varied from 0 (all months) to 1127 (August). Average value of C for 159 countries varied from 70 (June) to 120 (August). Value of CV changes from 141% (July) to 174% (June).

Range of change of mean monthly values of D for 159 countries varied from 0 (all months) to 19.0 (July). Average value of D for 159 countries varied from 1.3 (July) to 1.5 (August). Value of CV changes from 149% (August) to 189% (June and July). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint follows from Fig. 3 the highest growth rate of the monthly average values of D depending on C was observed in June, the smallest -in August (the corresponding values of the linear regression coefficients).

In Table 4 data about Covid-19 mean monthly values of infection (C) and deaths (D) cases in summer 2021 (per 1 million population) and ranking of Georgia, Armenia, Azerbaijan, Russia and Turkey by these parameters (in brackets) among 159 countries with population ≥ 1 million inhabitant are presented. The corresponding values of the deaths coefficient (DC) are also given here.

In particular, as follows from this Table, mean monthly values of C for 5 country changes from 6.8 (Azerbaijan, June, 110 place between 159 country) to 1127.1 (Georgia, August, 1 place between 159 country).

Mean monthly values of D for 5 country changes from 0.15 (Azerbaijan, July, 101 place between 159 country) to 13.74 (Georgia, August, 1 place between 159 country). Mean monthly values of DC for 5 country changes from 0.63% (Azerbaijan, July) to 3.10 (Russia, June).

Note that the mean values of DC (%) in summer 2021 are: Georgia -1.28, Armenia -2.08, Azerbaijan -0.80, Russia -3.36, Turkey -0.80, World -1.79 (according to data from Table 1 and 2). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint

The daily death from Covid-19 from June 1, 2021 to August 31, 2021 changes from 7 to 79 (Fig. 4) . A comparison between the daily mortality from Covid-19 in Georgia in summer 2021 with the average daily mortality rate in 2015-2019 shows (Fig. 5) , that the largest share value of D from mean death in 2015-2019 was 66.0 % (26.08.2021 and 31.08.2021), the smallest 6.0 % (09.07.2021). The most share of mean daily mortality from Covid-19 of mean daily mortality in 2015-2019 from October 2020 to August 2021 in August was observed -42.8% (Fig. 7 ).

Results of the statistical analysis of the daily and decade data associated with New Coronavirus COVID-19 pandemic in Georgia from June 1, 2021 to August 31, 2021 in Tables 5-7 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint

Mean decade values of deaths coronavirus-related cases (Fig. 9 ) varies from 13 (1 Decades of July) to 69 (3 Decade of August).

Mean decade values of infection rate coronavirus-related cases (Fig. 10) varies from 2.51 % (2 Decade of June) to 10.88 % (2 Decade of August).

Time changeability of the daily values of C, R, D and I are satisfactorily described by the tenth order polynomial ( Table 6 , Fig. 11-13 ). For clarity, the data in Fig. 12 are presented in relative units (%) in relation to their average values. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Note that from Fig. 11 and 12 , as in [7, 8] , clearly show the shift of the time series values of R and D in relation to C.

In Fig. 14-16 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint Table 7 are presented. As follows from this Table there was an increase in average monthly values of C, R and I from June to August. Value of D decrease from June to July, and further increase to August. The values of V(C), V(R), V(D) and V(I) changes as follows: V(C)from -9.6 cases/day (June) to +74.8 cases/day (July); V(R) -from -11.9 cases/day (June) to +113.0 cases/day (August); V(D) -from -0.5 cases/day (June) to +1.7 cases/day (August) and V(I) -from -0.01 %/day (June) to +0.16 %/day (July). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint In Fig. 17 data about connection of 14-day moving average of deaths cases due to COVID-19 in Georgia with 14-day moving average of infection rate from December 18, 2020 until August 31, 2021 are presented. As follows from Fig. 17 , in general, with an increase of the infection rate is observed increase of deaths cases. However, (Fig. 18 ), in August 2021 (I=10.4%), the average death rate from coronavirus (D=51.2) was more as in December 2020 (I=19.2, D= 40.0). This fact needs special analysis. It is possible that this is due to the predominance of the delta variant of the coronavirus in Georgia in the summer of 2021.

December 2020 to August 2021. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint Using the data in Fig. 18 , a linear regression graph between the monthly mean values of D and I is obtained (Fig. 19 ). As follows from Fig. 19 , in general a high level of linear correlation between these parameters is observed. This Fig. also clearly demonstrates the anomaly high mortality from coronavirus in August 2021 with relatively low value of infection rate in comparison with December 2020, which reduces the level of correlation between D and I.

As noted in [7, 8] and above ( Fig. 11 and 12) , there is some time-lag in the values of the time series R and D with respect to C. An estimate of the values of this time-lag is given below (Fig. 20) . 

In Fig. 21-24 and Table 8 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint Comparison of real and calculated predictions data of C, D and I shown, that two-week daily and mean two-week real values of C, D and I practically falls into the 67% -99.99% confidence interval of these predicted values for the specified time periods (Fig.  21 -24, data from https://www.facebook.com/Avtandil1948/, Table 8 ). Table 9 . Change in the forecast state of C, D and I in relation to the pre-predicted one according to Table 1 scale [6] .

In the future, it is planned to continue regular similar studies for Georgia in comparison with neighboring and other countries, in particular, checking the temporal representativeness of monthly interval statistical forecasting of the new coronavirus infection COVID-19, taking into account data on the periodicity, clarification of connections between various parameters of this infection, the influence of bioclimatic factors on this infection, etc. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 12, 2021. ; https://doi.org/10.1101/2021.09.08.21263265 doi: medRxiv preprint

we started monthly forecasting of C, D and I values. An interval forecast check of confirmed COVID-19 cases, deaths and infection rates in Georgia from 01

Report of the National Center for Disease Control & Public Health. The Fourth Revision

Report of the National Center for Disease Control & Public Health. The Fifth Revision

Report of the National Center for Disease Control & Public Health. The Six review

Twenty weeks of the pandemic of coronavirus Covid-19 in Georgia and neighboring countries

Analysis of twenty-week time-series of confirmed cases of New Coronavirus COVID-19 and their simple short-term prediction for Georgia and neighboring countries

The Statistical Analysis of Daily Data Associated with Different Parameters of the New Coronavirus COVID-19 Pandemic in Georgia and their Short-Term Interval Prediction from

The Statistical Analysis of Daily Data Associated with Different Parameters of the New Coronavirus COVID-19 Pandemic in Georgia and their Short-Term Interval Prediction in Spring 2021

Data Analysis of Covid-19 Pandemic and Short-Term Cumulative Case Forecasting Using Machine Learning Time Series Methods

Data Analysis and Forecasting of COVID-19 Pandemic in Kuwait

Preliminary Analysis of Excess Mortality in India During the Covid-19 Pandemic

Monitoring and Forecasting the SARS-Covid-19 Pandemic in France

Dynamics of COVID-19 Pandemic in India and Pakistan: A Metapopulation Modelling Approach

Predicting the Number of Reported and Unreported Cases for the Covid-19 Epidemics in China

Predictions of Covid-19 Dynamics in the UK: Short-Term Forecasting and Analysis of Potential Exit Strategies

Gradient-Based Grey Wolf Optimizer with Gaussian Walk: Application in Modelling and Prediction of the COVID-19 Pandemic

A Time Series-Based Statistical Approach for Outbreak Spread Forecating: Application of COVID-19 in Greece

Panel Forecasts of Country-Level COVID-19 Infections

Methoden der korrelations -und regressions analyse. -Ein Leitfaden für Ökonomen. Verlag Die Wirtshaft Berlin

Time-series