Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1, 381-392 2025 Publisher: Learning Gate DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate © 2025 by the authors; licensee Learning Gate History: Received: 26 November 2024; Revised: 30 December 2024; Accepted: 6 January 2025; Published: 8 January 2025 * Correspondence: omarin@unfv.edu.pe Prediction of deaths from COVID-19 with the modified logistic model, in Peru Marín Machuca, Olegario1,2,3*, Vargas Ayala, Jessica Blanca4, Pérez Ton, Luis Adolfo5, Chinchay Barragán, Carlos Enrique6, Rojas Rueda, María del Pilar7, Huaranja Montaño, Max Alejandro8, Sernaqué Auccahuasi, Fernando Antonio9 1Professional School of Food Engineering, Faculty of Oceanography, Fisheries, Food Sciences and Aquaculture, Federico Villarreal National University, Lima 15074, Peru; omarin@unfv.edu.pe (M.M.O.). 2Graduate University School, Federico Villarreal National University, Lima 15001, Peru. 3Environmental Sustainability Research Group (GISA), Lima 15001, Peru. 4Academic Department of Aquaculture, Faculty of Oceanography, Fisheries, Food Sciences and Aquaculture, Universidad Nacional Federico Villarreal, Lima 15074, Peru; jvargas@unfv.edu.pe (V.A.J.B.). 5,6Professional School of Food Engineering, Faculty of Fisheries and Food Engineering, Universidad Nacional del Callao. Callao, Peru; laperezt@unac.edu.pe (P.T.L.A.) cchinchayb@unac.edu.pe (C.B.C.E.). 7Professional School of Human Medicine, Norbert Wiener University, Lima, Peru; Maria.Rojasr@wiener.edu.pe (R.R.M.D.P.). 8University Graduate School (EUPG), Federico Villareal National University. Lima, Peru; 2022032509@unfv.edu.pe (H.M.M.A.). 9Faculty of Geographical, Environmental and Ecotourism Engineering, Federico Villarreal National University, Lima 15082, Peru; fsernaque@unfv.edu.pe (S.A.F.A.). Abstract: COVID-19 is a public health millions of deaths since the end problem that has had an international impact that has led to of 2019, and the Peruvian population was no stranger to this situation. Therefore, the following investigation was conducted to correlate mortality from COVID-19, estimate the critical time (days) for the maximum rate of estimated deceased people, and validate the reliability of the models. Data on people who died from COVID-19 up to February 27, 2023, were considered, with which the pandemic dispersion was carried out, arriving to determine that they describe a sigmoidal logistic dispersion, an event that was mathematically modeled using the predictive logistic equation N=M⁄((1+A×e^(-k×t))). Using this predictive mathematical model, the number and rate of deaths among people with COVID-19 in Peru were determined. In addition, the critical time (t_c) was estimated, whose value was t_c=396 days for the maximum rate 〖((dN ̂)⁄dt)〗_máx=484.7450 people/day, and the date on which the maximum rate of people who died from COVID-19 was April 15, 2021. The Pearson correlation coefficient between the time elapsed (t) and the number of deceased people (N) in Peru, based on 32 cases, turned out to be r=-0.89085; determining that the relationship is real, that there is a non-significant difference, that the predictive model has a high estimate of the correlated data, that there is a " very strong correlation " between the time elapsed (t) and the number of deceased people (N), and that 79.4% of the variance in N is explained by t; for people who died from COVID-19 in Peru. Keywords: COVID-19, Deceased in Peru, Estimate, Logistics modeling, Validation. 1. Introduction Coronavirus disease 2019 (COVID-19) is a person-to-person respiratory disease virus that was first identified in late 2019 during an outbreak investigation in Wuhan, China. It is currently known as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and is responsible for causing COVID- https://orcid.org/0000-0002-0515-5875 382 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate 19. It remains suspended in the air in open environments and travels over large distances due to atmospheric turbulence. and remains viable for less than 3 hours [1], whereas in closed environments, they are deposited superficially, and their activity continues. Experimental studies have shown that the virus can persist for at least three hours in aerosols, 24 hours on cardboard, and up to 72 hours on plastic or stainless-steel surfaces. The virus has been detected in the gastrointestinal tract, feces, saliva, and urine, and its potential transmission routes need to be evaluated in the near future [2]. On March 18, 2020, World Health Organization (WHO) and its partners launch the “Solidarity” trial, an international clinical trial that aims to generate robust data from around the world to find the most effective treatments against COVID-19. Today, it is accepted that contagion among asymptomatic subjects has been the main cause of the extension and propagation of the SARS-CoV-2 pandemic [3], and its spread speed was higher than that of SARS-CoV in 2003 [4]. The possible risk factors included age, sex, smoking status at the time of infection, chronic obstructive pulmonary disease, coronary disease, diabetes, arterial hypertension, carcinoma, chronic kidney disease, and other comorbidities. In a univariate study, the following variables were significantly associated with higher mortality: age, coronary disease, diabetes, and hypertension [5]. On March 6, 2020, the first case of coronavirus was reported in Peru and since then the progress of the pandemic has been evaluated by the Ministerio de Salud del Peru (MINSA) and reported to the public, with 59.2% of cases in adults, that the highest cumulative incidence rates correspond to the group of adults and the elderly; that the analysis of the ratio of incidence rates is 12 times more in adults compared to the rate of children and so far in the pandemic in Peru, cough, fever, general malaise and sore throat are the most frequent symptoms [6]. The evolution of the death of people in Peru by COVID-19 until February 27, 2023, in terms of determining the highest mortality, statistical mathematical modeling, critical time (days), the rate at which the death developed, validation of the estimated data, along with other Peruvian and global public health indicators, constitutes a true problem of prevention, which surely serves as referential data to face similar problems of deaths [4]. As of February 27, 2023, approximately 219,000 deaths from the pandemic have been registered in Peru according to the Chinese CDC series, with a total of 1,023 deaths among confirmed cases (44,672) and a gross lethality of 2.3 %. The age group older than 80 years had the highest lethality among all the age groups (14.8 %). Patients without comorbidities had a case fatality rate of 0.9 % compared to patients with comorbidities who had higher rates: 10.5 % for those with cardiovascular disease, 7.3 % for diabetes, 6.3 % for chronic respiratory disease, 6.0 % for hypertension, and 5.6 % for cancer [7]. A mathematical model of the logistic type is a tool that helps us analyze and estimate the problems caused by diseases. Its objective is to describe, explain, and predict phenomena such as epidemics in specific geographical areas, aiming to understand the dynamics of dispersal and, in this case, mortality due to the disease in various scenarios [8]. These models, including the long short-term memory (LSTM) model for predicting the number of weekly and daily COVID patients [9] the age-stratified mathematical model for describing differences in biological susceptibility to the disease based on age and evaluating the morbidity rate per infection [10, 11] and the exponential model for estimating the growth of new COVID cases [12], have demonstrated good accuracy in forecasting the behavior of the infection. Marín-Machuca, et al. [7] mention that the modeling for COVID-19 was based on determining the relationship between the variation in the number of reported cases (𝑑𝑁) and the variation in the time elapsed (𝑑𝑡), called velocity of cases reported with respect to the time of what happened in China. To find a relationship that adequately estimated the COVID-19 infections, a corresponding predictive logistic model was obtained. Manrique-Abril, et al. [13] mentioned that mathematically modeling cases and phenomena that contain the exponential function of the form 𝑁 = 𝑀 (1 + 𝐴𝑒𝑘×𝑡)⁄ …(1); where "𝑀" is a maximum possible quantity, "𝐴" is a pre-exponential quantity, "𝑘" is a constant of proportionality, "𝑡" is the elapsed time of contagion (days) and "𝑁" is the number of deaths. It is 383 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate induced to evaluate the values of the constant of proportionality (k) or rate of change of the contagious phenomenon and 𝐴 it is a pre-exponential factor. The objectives of the present study were to analyze the mortality behavior in Peru due to the SARS-CoV-2 pandemic, which caused COVID-19, compare the representations between deceased and estimated deceased, estimate the critical time (days) for the rate number of estimated deceased people, and statistically validate the reliability of the models. 2. Methodology 2.1. Data Sources The methodology used was based on the specific growth constant (k), where the conditions of the process will exercise restrictions on deaths from COVID-19 in Peru, bearing in mind that the constant k will decrease as infections increase, and starting from the fact that the k of the deceased only depends on the number of people and not on time-dependent mechanisms, such as non-seasonal phenomena, arriving to determine a logistic equation, in which its solution is a logistic function and whose purpose is to understand the number of deaths and why not make predictions regarding future behavior. The stages covered were: 1) the problem of modeling the number of deaths as a function of time, 2) formulating and choosing, through the dispersion of the data, the logistic model, 3) determining the model, analyzing it, and drawing mathematical conclusions, and 4) making predictions (estimates) about the number of deaths from COVID-19 in Peru and the rate at which deaths have occurred. Considering that the mathematical model is never a completely accurate representation and that it is only an idealization that simplifies the reality of those who died from COVID-19 in Peru, it is accurate enough to promote valuable conclusions and relevant discussions. As of February 27, 2023; 219,431 people have died from the coronavirus (SARS-CoV-2) in Peru, causing the disease COVID-19, which initially originated in the city of Wuhan (China) and has spread to every country in the world [7]. The accumulated cases of deceased people in our country based on the elapsed time (days) are presented in Table 1. Table 1. Statistical data on the number of people who died from COVID-19 in Peru based on the time elapsed (days). Date Time. t (Days) N (Accumulated cases) 03/16/2020 0 13 04/07/2020 22 498 05/15/2020 60 12.557 06/22/2020 98 34.757 07/31/2020 137 57.462 08/24/2020 161 71.839 09/07/2020 175 77.609 10/15/2020 213 85.855 11/21/2020 250 90.072 12/29/2020 288 94.448 02/06/2021 327 109.551 03/16/2021 365 135.306 04/23/2021 403 164.069 05/31/2021 441 186.385 07/09/2021 480 195.385 08/14/2021 516 198.605 09/21/2021 554 200.208 10/29/2021 593 201.372 12/07/2021 631 202.653 01/14/2022 668 204.280 02/20/2022 705 210.189 03/30/2022 743 212.290 05/07/2022 781 212.956 06/14/2022 819 213.405 384 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate 07/23/2022 858 214.111 08/30/2022 896 215.816 09/07/2022 904 216.069 11/10/2022 938 216.830 11/05/2022 963 217.030 12/25/2022 1.013 218.072 01/11/2023 1.030 218.455 02/27/2023 1.077 219.431 Source: Sistema Informático Nacional De Defunciones [14]. Figure 1. Representation of the number of people who died from COVID-19 in Peru based on the elapsed time (days). In Figure 1, the evolution of accumulated cases is plotted as a function of time, showing that the accumulated cases increase with time. 2.2. Statistical Treatment The model was evaluated according to the following criteria: 1. R-Squared Score. According to Lupón, et al. [15] is a measure that quantifies the relationship between the dependent variable and the mathematical model. The points around the regression are quantized in the closed interval of 0 to 100%. 𝑅2 = 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝑒𝑥𝑝𝑙𝑎𝑖𝑛𝑒𝑑 𝑏𝑦 𝑡ℎ𝑒 𝑚𝑜𝑑𝑒𝑙 𝑇𝑜𝑡𝑎𝑙 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 385 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate 2. Adjusted R-Squared Score. t is a modified model of R2, it serves as an indicator to quantify how close the points of the curve are. where n is the data size and k is the number of independent variables in the regression equation [15]. 𝑅𝑎𝑗𝑢𝑠𝑡𝑒𝑑 2 = 1 − (1 − 𝑅2) (𝑛 − 1) 𝑛 − (𝑘 + 1) 3. Mean Absolute Error (MAE). The mean absolute error is the average magnitude between the data predicted by the model and the values obtained from the real world, in this measure each pair of data that is compared has the same weight. Where n is the number of data, is the theoretical value and �̂�𝑗 is the value obtained from the real data [16]. 𝑀𝐴𝐸 = 1 𝑛 ∑|𝑁𝑗 − �̂�𝑗| 𝑛 𝑗=1 4. Mean Square Error (MSE). It is used to measure the efficiency of the model. It is calculated by means of the difference between the theoretical value and the real value, square it, all the values are positive, and it is characterized by the fact that it applies better when the differences between the compared values are greater [16]. 𝑀𝑆𝐸 = 1 𝑛 ∑(𝑁𝑖 − �̂�𝑖) 2 𝑛 𝑖=1 5. Root Mean Square Error (RMSE). Calculates the errors of the predicted values. These values are the residuals between the predicted value and the actual value, it explains how close the actual values are to the model [16]. 𝑅𝑀𝑆𝐸 = √ 1 𝑛 ∑(𝑁𝑖 − �̂�𝑖) 2 𝑛 𝑖=1 Hernández and Fenández [17] mentioned that in the statistical treatment of correlated data, there is a Pearson correlation coefficient, which has a relative interpretation and indicates the magnitude of the relationship between the dependent and independent variables, and the sign only indicates the direction of the relationship. To validate the models obtained, using the correlation and determination coefficients, a significance test was performed 𝑟, to determine if this value represents a real relationship between the two variables. The standard error of 𝑟 was calculated using the expression: 𝑡𝑐𝑎𝑙 = |𝒓| √1−𝒓2 × √𝑁 − 2… (2) Comparing those 𝑡 of Student, calculated (𝑡𝑐𝑎𝑙) and table (𝑡𝑡𝑎𝑏), the relationship between the elapsed time 𝑡 (days) and the number of deceased people (𝑁), the degree of difference and the estimate of the predictive model was determinate. 2.3. Correction of the Estimated Logistic Model Marín-Machuca, et al. [18] mention that the models that present an effect called "loop", similar to a hysteresis phenomenon, as can be seen in Figure 2, a factor of correction (“period”) for the independent variable t. For this purpose, parameter T was evaluated, which acts as the period given by the expression (3): 𝑇 = ∑ (𝑁−𝑓(𝑡)) 𝑡𝑛 𝑡1 𝑛(∆𝑡) … (3) 386 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate 3. Results To model the behavior of the number of people who died from COVID-19 in Peru, we based ourselves on the Empirical Modeling theory [19] on the number of people who died (𝑁), as a function of the elapsed time, 𝑡 (days). Determining the behavior (figure 1) of the statistical data (Table 1) of the number of people who died from COVID-19, it has been considered that the model is logistic, of the type of equation (1). The method of calculating 𝑀 is by considering three independent values and their corresponding dependent values from Table 1. Bronshtein and Semendiaev [19] mentioned that to obtain the maximum values (𝑀) and for this type of logistic model, the first value (𝑃) must be precisely at the moment in which the behavior presents an inflection point, the second value (𝑄) is the last data and the third value (𝐼) is an intermediate value between the values 𝑃 and 𝑄. This intermediate value is the average of the first and last values. The following formula was then applied: 𝑀 = 𝑃×𝑄−𝐼2 𝑃+𝑄−2𝐼 …(4) First value: 𝑡1 = 365 𝑑𝑎𝑦𝑠, corresponds to: 𝑃 = 135,306 deaths Second value: 𝑡2 = 1,077 𝑑𝑎𝑦𝑠, corresponds to: 𝑄 = 219,431 deaths Third value: 𝑡3 = 365+1 077 2 = 721 𝑑𝑎𝑦𝑠, corresponds to: 𝐼 = 211,416 deaths Now, we substitute into equation (4). 𝑀 = 135,306×219,431−211,4162 135,306+219,431−2(211,416) = 220,374 deaths The model 𝑁 = 𝑀 1+𝐴×𝑒𝐾×𝑡 can be written �̂� = 220,374 1+𝐴×𝑒𝑘×𝑡 Applying the method of least squares to the expression ln ( 220,374 𝑁 − 1) = 𝐴 + 𝑘 × 𝑡, we obtain the prediction or estimation model. �̂� = 220,374 1+32.5421×𝑒−0.0088×𝑡…(5) With a correlation coefficient 𝑟 = −0.89085. Deriving equation (5), the equation for the rate of infected people is obtained and expressed by equation (6). 𝑑𝑁 𝑑𝑡 = 63,098.55672×𝑒−0.0083×𝑡 (1+32.5421×𝑒−0.0083×𝑡)2…(6) Deriving equation (6) and equaling it to zero, it is possible to determine the critical time (𝑡𝑐) for which the rate of the infected people is maximum. 𝑡𝑐 = − 1 𝑘 × ln (1 𝐴⁄ )… (7) Then: 𝑡𝑐 = 396 𝑑𝑎𝑦𝑠 and the maximum velocity is ( 𝑑𝑁 𝑑𝑡 )𝑚á𝑥 = 484.7450 𝑑𝑒𝑎𝑡ℎ𝑠/𝑑𝑎𝑦 Scheduling the process of people who died from COVID-19 in Peru; April 15, 2023, was the date on which there was the maximum rate of deaths. The number of estimated deaths in Peru due to COVID-19 is determined by equation (3) and is represented in figure 2. The rate of estimated deaths due to COVID-19 is determined by equation (6) and is represented in Figure 3. In addition, Table 2 presents the data on time, accumulated deaths, estimated accumulated deaths, and the rate of estimated deaths due to COVID-19 in Peru. 387 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate Table 2. Time data, accumulated cases of deaths, estimated deaths, and rate of deaths estimated by COVID-19 in Peru. Time. t (Days) N (Accumulated deaths) �̂� 𝒅𝑵 𝒅𝒕⁄ (𝒑𝒆𝒐𝒑𝒍𝒆/𝒅𝒂𝒚) �̂�𝒎 0 13 6.570 56.0840 11.521 22 498 7.923 67.2054 13.841 60 12.557 10.913 91.2682 18.900 98 34.757 14.953 122.6400 25.578 137 57.462 20.506 163.6321 34.473 161 71.839 24.786 193.5533 41.109 175 77.609 27.628 212.6135 45.413 213 85.855 36.769 269.5371 58.730 250 90.072 47.847 329.5851 73.857 288 94.448 61.542 390.2665 91.168 327 109.551 77.840 442.9689 109.991 365 135.306 95.373 475.9832 128.352 403 164.069 113.704 484.2516 145.731 441 186.385 131.843 466.0206 161.338 480 195.385 149.263 423.7820 175.023 516 198.605 163.597 370.8539 186.208 554 200.208 176.523 309.0515 194.229 593 201.372 187.353 247.0015 201.197 631 202.653 195.688 192.8745 206.324 668 204.280 201.976 148.3674 210.064 705 210.189 206.773 112.2853 212.845 743 212.290 210.465 83.2659 214.942 781 212.956 213.190 61.1521 216.466 819 213.405 215.184 44.5933 217.568 858 214.111 216.666 32.0762 218.379 896 215.816 217.707 23.1803 218.945 904 216.069 217.886 21.6401 219.042 938 216.830 218.524 16.1383 219.389 963 217.030 218.887 12.9943 219.582 1.013 218.072 219.414 8.4092 219.865 1.030 218.455 219.547 7.2495 219.936 1.077 219.431 219.826 7.8061 220.085 Note: 𝑁: Estimated deaths 𝑁𝑚: Modified estimate of deaths. 388 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate Figure 2. Representation of the number of deaths and the estimated number of deaths from COVID-19 in Peru based on the time elapsed (days). In Figure 2 The number of deaths and the number of estimated deaths from COVID-19 in Peru are represented based on the time elapsed, the accumulated cases and the mathematical model used. 389 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate Figure 3. Rate of the number of people estimated to have died (people/day) due to COVID-19 in Peru based on the time elapsed (Days). In Figure 3, the rate of the number of estimated deaths (people/day) by COVID-19 in Peru is graphed as a function of the time elapsed, and the maximum point in time of 396 days was observed. 390 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate Figure 4. Representation of accumulated (𝑁), estimated (�̂�), and modified estimate (�̂�𝑚) deaths. Pearson 𝑁 correlation coefficient “ 𝑟” for the elapsed time t (days) and the number of people who died (𝑁) in Peru from COVID-19 based on 32 cases was 𝑟 = −0.89085. The standard error of 𝑟 was calculated using the equation (1): 𝑡𝑐𝑎𝑙 = |−0.89085| √1 − (−0.89085)2 × √32 − 2 = 23.64201 and 𝑡𝑡(30;0.95) = 1.7033 Because 𝑡𝑐𝑎𝑙 = 23.64201 is greater than 𝑡𝑡𝑎𝑏 = 1.7033; it is concluded that the relationship between time, 𝑡(days) and the number of deceased people (𝑁) is real; therefore, there is a non-significant difference and that the predictive model (equation 2) has a high estimate of the correlated data and that there is a "very strong correlation" between the time elapsed (𝑡) and the number of deceased people (𝑁) and that the 79.4 % of the variance in 𝑁 is explained by 𝑡; for the number of people who died from COVID-19 in Peru. In addition, the parameter 𝑇, which acts as the period, was calculated, resulting in, 𝑇 = ∑ (𝑁 − 𝑓(𝑡))𝑡=593 𝑡=98 15(593 − 98) = ∑ (𝑁 − �̂�)𝑡=593 𝑡=98 15(593 − 98) = 490,833 7,425 = 66.1055 �̂� = 220,374 1+47.38779×𝑒−0.00855×(𝑡+66.1055)…(6) To compare the logistic model and the modified logistic model, three equations were used to quantify the errors, namely MAE (Mean Absolute Error), MSE (Mean Squared Error), and RMSE (Root Mean Squared Error). In all cases, it was found that the modified logistic model has lower error compared to the logistic model, indicating a better fit of the data to the proposed model. 391 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate Table 3. Parameter values. No Parameter Logistic model Modified logistic model 1 R-squared score 79.40% 97.59 % 2 Adjusted R-squared score 84.25% 84.25% 3 Mean absolute error (MAE) 19.318 9.514 4 Mean square error (MSE) 755’092,668 178’640,321 5 Root mean square error (RMSE) 27.479 13.366 4. Results and Discussion The mathematical model (equation 3) to estimate the number of people who died from COVID-19 in Peru turned out to be acceptable, reaching a Pearson correlation coefficient of , 𝑟 = −0.89085 coinciding with what was reported by Florencio [8]. From the mathematical model of rate (equation 4), it is estimated that the critical time (𝑡𝑐) is 396 días, which corresponds to the maximum rate of people estimated to have died from COVID-19 in Peru of 484.7450 𝑝𝑒𝑜𝑝𝑙𝑒/𝑑𝑎𝑦, whose scheduled date was April 15, 2021, coinciding with what was reported by Marín-Machuca, et al. [7] and Marín-Machuca, et al. [7]. The value of the correction factor for the independent variable is 66.1055 days, whose predictive mathematical model (equation 6), the constant of proportionality (𝑘 = −0.0088) and the coefficients of correlation (𝑟 = −0.9879) and determination (𝑟2 × 100 = 97,59%) are of great importance for analyze and estimate data on epidemiological and pandemic phenomena, observable in figure 3; coinciding with what was mentioned by Hernández and Fenández [17]. The theory of Bronshtein and Semendiaev [19] can be applied without difficulty, if it is considered that the processes show behavior that will not always ascend or will not always descend. The logistic model provides excellent prediction, with an error of less than 7% on a weekly basis and less than 12% monthly. This aligns with the findings of Arora, et al. [9] who reported an error of less than 8% on a weekly basis and less than 14% monthly. The number of deaths due to COVID-19 follows a behavior consistent with a logistic model, which helps adjust the trend limit of the epidemic. This corresponds to the findings reported by Wang, et al. [20]. 5. Conclusion Logistic (factual) model can generally be applied as rigorously as possible to pandemic and epidemiological phenomena with high resolution and a high degree of estimation of real data. Statistically, it has been determined that the correlation coefficient of equation 3 has a "very strong negative correlation" between the number of infections by COVID-19 and the time elapsed, 79.40% of the variance in 𝑁 is explained by 𝑡 for the number of deaths in Peru from COVID-19. To obtain a better estimate of the predictive model, it is recommended that the statistical data, in terms of the dependent variable (consisting of the number of people infected by COVID-19). Transparency: The authors confirm that the manuscript is an honest, accurate, and transparent account of the study; that no vital features of the study have been omitted; and that any discrepancies from the study as planned have been explained. This study followed all ethical practices during writing. Copyright: © 2025 by the authors. This open-access article is distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). References [1] X.-X. Liu et al., "Effects of air pollutants on occurrences of influenza-like illness and laboratory-confirmed influenza in Hefei, China," International Journal of Biometeorology, vol. 63, pp. 51-60, 2019. https://creativecommons.org/licenses/by/4.0/ 392 Edelweiss Applied Science and Technology ISSN: 2576-8484 Vol. 9, No. 1: 381-392, 2025 DOI: 10.55214/25768484.v9i1.4135 © 2025 by the authors; licensee Learning Gate [2] N. Van Doremalen et al., "Aerosol and surface stability of SARS-CoV-2 as compared with SARS-CoV-1," New England Journal of Medicine, vol. 382, no. 16, pp. 1564-1567, 2020. https://doi.org/10.1056/NEJMc2004973 [3] Y.-R. Guo et al., "The origin, transmission and clinical therapies on coronavirus disease 2019 (COVID-19) outbreak– an update on the status," Military Medical Research, vol. 7, pp. 1-10, 2020. https://doi.org/10.1186/s40779-020- 00240-0 [4] A. Ruiz-Bravo and M. Jiménez-Valera, "SARS-CoV-2 y pandemia de síndrome respiratorio agudo (COVID-19)," Ars Pharmaceutica (Internet), vol. 61, no. 2, pp. 63-79, 2020. [5] F. K. Ho et al., "Modifiable and non-modifiable risk factors for COVID-19, and comparison to risk factors for influenza and pneumonia: Results from a UK Biobank prospective cohort study," BMJ open, vol. 10, no. 11, p. e040402, 2020. https://doi.org/10.1136/bmjopen-2020-040402 [6] A. Revollé, "Coronavirus in Peru: This is how the pandemic is evolving in the country. Diario La República," Retrieved: https://data.larepublica.pe/envivo-casos-confirmados-muertes-coronavirus-peru/. 2020. [7] O. Marín-Machuca, A. W. Zambrano-Cabanillas, E. G. García-Talledo, J. I. Ortiz-Guizado, D. E. Vivas-Ruiz, and O. Marín-Sánchez, "Mathematical modeling of the epidemiological behavior of the COVID-19 pandemic in China," The Biologist, vol. 18, no. 1, 2020. https://doi.org/10.24039/rtb2020181473 [8] C. Florencio, "Statistical calculations on a closed SIR model extrapolating data from the current Coronavirus outbreak to a Mexican population scenario," Magdalena Contreras Mayor's Office, 2020. [9] P. Arora, H. Kumar, and B. K. Panigrahi, "Prediction and analysis of COVID-19 positive cases using deep learning models: A descriptive case study of India," Chaos, Solitons & Fractals, vol. 139, p. 110017, 2020. https://doi.org/10.1016/j.chaos.2020.110017 [10] H. H. Ayoub, G. R. Mumtaz, S. Seedat, M. Makhoul, H. Chemaitelly, and L. J. Abu-Raddad, "Estimates of global SARS-CoV-2 infection exposure, infection morbidity, and infection mortality rates in," Global Epidemiology, vol. 3, p. 100068, 2021. https://doi.org/10.1016/j.gloepi.2021.100068 [11] H. H. Ayoub et al., "Characterizing key attributes of COVID-19 transmission dynamics in China’s original outbreak: Model-based estimations," Global Epidemiology, vol. 2, p. 100042, 2020. https://doi.org/10.1016/j.gloepi.2020.100042 [12] W. Koczkodaj et al., "1,000,000 cases of COVID-19 outside of China: The date predicted by a simple heuristic," Global Epidemiology, vol. 2, p. 100023, 2020. https://doi.org/10.1016/j.gloepi.2020.100023 [13] F. G. Manrique-Abril, C. A. Agudelo-Calderon, V. M. González-Chordá, O. Gutiérrez-Lesmes, C. F. Téllez-Piñerez, and G. Herrera-Amaya, "Modelo SIR de la pandemia de Covid-19 en Colombia," Revista de Salud Pública, vol. 22, pp. 123-131, 2023. https://doi.org/10.15446/rsap.v22n2.85977 [14] Sistema Informático Nacional De Defunciones, "Number of people who died from COVID-19 in Peru. Ministerio de Salud," Retrieved: https://www.minsa.gob.pe/defunciones/. 2023. [15] J. Lupón et al., "Biomarker-assist score for reverse remodeling prediction in heart failure: The ST2-R2 score," International Journal of Cardiology, vol. 184, pp. 337-343, 2015. https://doi.org/10.1016/j.ijcard.2015.02.019 [16] C. J. Willmott and K. Matsuura, "Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance," Climate Research, vol. 30, no. 1, pp. 79-82, 2005. https://doi.org/10.3354/cr030079 [17] R. Hernández and C. Fenández, B. P., Research methodology. In Self-training notebooks on social participation. CAPS 002, 2018. [18] O. M. Marín-Machuca, J. B. V. Ayala, U. M. Sánchez, F. A. A. Zambrano, E. E. L. K. Prado, and O. M. Sánchez, "Mathematical modeling of COVID-19 mortality in China," Cátedra Villarreal, vol. 8, no. 1, pp. 35-43, 2020. [19] I. Bronshtein and K. Semendiaev, "Manual de matemáticas," p. 694, 2018. [20] P. Wang, X. Zheng, J. Li, and B. Zhu, "Prediction of epidemic trends in COVID-19 with logistic model and machine learning technics," Chaos, Solitons & Fractals, vol. 139, p. 110058, 2020. https://doi.org/10.1016/j.chaos.2020.110058 https://doi.org/10.1056/NEJMc2004973 https://doi.org/10.1186/s40779-020-00240-0 https://doi.org/10.1186/s40779-020-00240-0 https://doi.org/10.1136/bmjopen-2020-040402 https://data.larepublica.pe/envivo-casos-confirmados-muertes-coronavirus-peru/ https://doi.org/10.24039/rtb2020181473 https://doi.org/10.1016/j.chaos.2020.110017 https://doi.org/10.1016/j.gloepi.2021.100068 https://doi.org/10.1016/j.gloepi.2020.100042 https://doi.org/10.1016/j.gloepi.2020.100023 https://doi.org/10.15446/rsap.v22n2.85977 https://www.minsa.gob.pe/defunciones/ https://doi.org/10.1016/j.ijcard.2015.02.019 https://doi.org/10.3354/cr030079 https://doi.org/10.1016/j.chaos.2020.110058