#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Tue Jun 27 18:42:36 2023

@author: richardson
"""

import statistics as st
import pandas as pd
import numpy as np

WS = pd.read_csv('sampleData.csv', header=None)
bubba = np.array(WS)
#print(bubba)

xVals = [b[0] for  b in bubba]
yVals = [b[1] for b in bubba]
#print(xVals)
#print(yVals)
mx = st.mean(xVals)
my = st.mean(yVals)
sx = st.pstdev(xVals)
sy = st.pstdev(yVals)
xy = [a[0]*a[1] for a in bubba]
sumxy = sum(xy)
sumx = sum(xVals)
sumx2 = sum(a[0]**2 for a in bubba)
sumy = sum(yVals)
N = len(bubba)

# linear regression equation
a = (sumy*sumx2-sumx*sumxy)/(N*sumx2-sumx*sumx)
b = (N*sumxy-sumx*sumy)/(N*sumx2-sumx*sumx)
LRVals = [a+b*xVals[j] for j in range(N)]

#We write the equation
print("Our equation is y = "+str(round(a,3))+" + "+str(round(b,3))+"x")

# another way to make the linear
# regression vector of predicted 
# yVals
def LR(t):
    return a+b*t

LRvect = [LR(a) for a in xVals]
#print(LRVals)
#print(LRvect)
import matplotlib.pyplot as plt
plt.scatter(xVals,yVals, color='red')
xmin = min(xVals)
xmax = max(xVals)
xs = np.linspace(xmin,xmax,100)
ys = [LR(t) for t in xs]
plt.plot(xs, ys, color= 'blue')
plt.show()