"An approximate answer to the right problem is worth a good deal more than an exact answer to an approximate problem."
-John Tukey
Math modeling: Covid-19 daily cases simulation
Based on Johns Hopkins COVID-19 map registered daily cases, we could simulate the behavior of the daily reported cases. First of all, there is need to actually download the data directly from internet, after that just graph directly the data.
​
After that using the Excel tool "Solver" we adjust the data to a normal distribution equation, the adjustment will done by the reduction RMSE.
​
For the following examples the data shown will be the cases reported in USA from March 12, 2020 to April 24,2020.
​
​
​
![NOrmal.PNG](https://static.wixstatic.com/media/905dab_7f537a60a0574f6a8640b8f4941a7e2b~mv2.png/v1/fill/w_256,h_64,al_c,lg_1,q_85,enc_avif,quality_auto/NOrmal_PNG.png)
![NormalDistr.PNG](https://static.wixstatic.com/media/905dab_d69d3356a29543efae2102559efb11e2~mv2.png/v1/fill/w_528,h_288,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/NormalDistr_PNG.png)
![Table.PNG](https://static.wixstatic.com/media/905dab_5d32869f8f574c44ac936c939411c9eb~mv2.png/v1/fill/w_333,h_84,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/Table_PNG.png)
The graph shown describes the behavior of the confirmed cases over time. The table shown on the left side of the graph displays the coefficents obtained by Solver, these coefficients have different meanings, the p value approximated the maximum cases registered in the observation represented by μ. Finally, the value σ is the value of the standard deviation of the distribution.
​
This kind of representation helps to visualized the posible behavior of the pandemy and shows an approximated of what is the tentative date of how long the pandemy might be.
​
Showing up next there are 3 graphs showing the model and real confirmed cases updated (April 30,2020) for Mexico, USA and the world in general with their respective coefficients.
Mexico
![StatsMex.PNG](https://static.wixstatic.com/media/905dab_8a1a0bf74df448f0ad2e6662c582afaa~mv2.png/v1/fill/w_213,h_131,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/StatsMex_PNG.png)
![CasesMEx.PNG](https://static.wixstatic.com/media/905dab_ad059fd40b6f4418aab00a6fdb4476e3~mv2.png/v1/fill/w_600,h_331,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/CasesMEx_PNG.png)
![varmex.PNG](https://static.wixstatic.com/media/905dab_02ff6235c7b4477ab1184544656edfd6~mv2.png/v1/fill/w_296,h_75,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/varmex_PNG.png)
United States of America
![Coef2USA.PNG](https://static.wixstatic.com/media/905dab_700b20c4cea646c89e6a5abb12388f81~mv2.png/v1/fill/w_199,h_128,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/Coef2USA_PNG.png)
![CasesUSA.PNG](https://static.wixstatic.com/media/905dab_7a657f71214148e48653cbcbb0fb4b95~mv2.png/v1/fill/w_598,h_320,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/CasesUSA_PNG.png)
![Coef1USA.PNG](https://static.wixstatic.com/media/905dab_05237c7e766c46668edb890e82faa4da~mv2.png/v1/fill/w_316,h_80,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/Coef1USA_PNG.png)
World
![CasesWorld.PNG](https://static.wixstatic.com/media/905dab_f26a56ebd8b846c8b80e154b7e58ac04~mv2.png/v1/fill/w_601,h_326,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/CasesWorld_PNG.png)
![StatsW.PNG](https://static.wixstatic.com/media/905dab_e6eb5dec994f4fe1aeb840c712236610~mv2.png/v1/fill/w_241,h_148,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/StatsW_PNG.png)
![VarW.PNG](https://static.wixstatic.com/media/905dab_321cedcfeed24dec8a0a67374cb11bc4~mv2.png/v1/fill/w_344,h_84,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/VarW_PNG.png)
Download the Excel file
If you like this analysis, try to download th csv I made for you, where you can play with the parameters and data.
What is going on right now?
References
Right now, i´m writing a code in order to automatized this tasks: obtain data and fit the data to the model