1. 3. cumulative covid-19 cases and
   4. attempt to predict ca. 2 months old wave with logistic curve
   6. 12.1.2021
   7. 0000.0001

import numpy as np import numpy.random as npr import matplotlib.pyplot as plt from scipy.optimize import leastsq import math as math import pandas as pd import matplotlib.pyplot as plt import matplotlib.ticker as ticker import locale from datetime import datetime, timedelta import matplotlib.dates as mdates from statsmodels.nonparametric.kernel_regression import KernelReg from scipy.signal import savgol_filter

from matplotlib.ticker import (MultipleLocator, FormatStrFormatter,

                              AutoMinorLocator)

def logistic4(x, A, B, C, D):

   """4PL lgoistic equation."""
   return ((A-D)/(1.0+((x/C)**B))) + D

def residuals(p, y, x):

   """Deviations of data from fitted 4PL curve"""
   A,B,C,D = p
   err = y-logistic4(x, A, B, C, D)
   return err

def peval(x, p):

   """Evaluated value at x with current parameters."""
   A,B,C,D = p
   return logistic4(x, A, B, C, D)

dfin = pd.read_csv('https://datahub.io/core/covid-19/r/countries-aggregated.csv', parse_dates=['Date']) countries = ['Finland'] dfin = dfin[dfin['Country'].isin(countries)]

selected_columns = dfin"Date", "Confirmed", "Recovered", "Deaths"

df2 = selected_columns.copy()

df2.to_csv (r'kovadata1.csv', index = True, header=True, sep=';')

df = pd.read_csv(r'kovadata1.csv', delimiter=';')

len1=len(df["Date"]) aktiv2= [None] * len1 for n in range(0,len1-1): aktiv2[n]=0

dates=df['Date'] rekov1=df['Recovered'] konf1=df['Confirmed'] death1=df['Deaths']

dailycases1= [0] * len1 dailydeaths1= [0] * len1

for n in range(1,(len1-1)): dailycases1[n]=konf1[n]-konf1[n-1] if (dailycases1[n]<0): dailycases1[n]=0

for n in range(1,(len1-1)): dailydeaths1[n]=death1[n]-death1[n-1] if (dailydeaths1[n]<0): dailydeaths1[n]=0

locale.setlocale(locale.LC_TIME, "fi_FI") dates_a = [datetime.strptime(d,'%Y-%m-%d').date() for d in dates]

dates_aaa = pd.date_range(dates_a[0], periods=len1, freq='1D') dates_aa=np.array(dates_aaa)

1. 1. try to predict covid-19 cases
   2. logistic distribution

1. 1. item to begin specimen

nalku1=20 ## 15 pv ennen 1. tapausta

print ("Alkupvm ", dates_a[nalku1])

konf2=konf1[nalku1:] dates_b=dates_a[nalku1:]

nayte_alku=0

1. 1. length of specimen to

nayte_loppu=60 ## 2 kk nayte_maara=nayte_loppu-nayte_alku ennuste_alku=0 ennuste_loppu=175 ennuste_maara=ennuste_loppu-ennuste_alku

x = np.linspace(0,nayte_maara,nayte_maara) x2=np.linspace(0,ennuste_maara,ennuste_maara) x3=np.linspace(0,len1,len1) x4=np.linspace(0,len1-nalku1,len1-nalku1)

dates_c=dates_a[nalku1:nalku1+nayte_maara]

dates_ddd=pd.date_range(dates_c[0], periods=ennuste_maara, freq='1D') dates_d=np.array(dates_ddd)

A,B,C,D = 0.5,2.5,8,7.3

1. y_true = logistic4(x, A, B, C, D)

y_meas=konf2[nayte_alku:nayte_loppu]

y_meas2 = savgol_filter(y_meas, 15, 2)

1. Initial guess for parameters

p0 = [0, 1, 1, 1]

1. plsq = leastsq(residuals, p0, args=(y_meas, x))

plsq = leastsq(residuals, p0, args=(y_meas2, x))

y2=peval(x2,plsq[0])

1. plt.plot(x,y_meas,'bo',x2,y2, "g--", x4,konf2,'r+')

plt.plot(dates_c,y_meas,'ro', dates_aa,konf1,'bx', dates_d,y2, 'g--')

1. plt.plot(dates_b,y_meas,'bo', x2,y2, "g--",dates_b,konf2,'r+')

plt.title('Kasautuneet koronavirustapaukset verrattuna logistiseen käyrään', fontsize=12)

1. plt.legend()

plt.legend(['Tutkitut tapaukset \n Jakso: alku 15 pv ennen 1. tapausta, \n siitä eteenpäin 60 päivää', 'Todetut tapaukset', 'Ennuste (logistinen käyrä)'], loc='upper left')

1. for i, (param, actual, est) in enumerate(zip('ABCD', [A,B,C,D], plsq[0])):
2. plt.text(10, 3-i*0.5, '%s = %.2f, est(%s) = %.2f' % (param, actual, param, est))

plt.show()

1. plt.savefig('logistic.svg')